{"paper_id":"5578fba2-74b5-4327-bd8f-8944f61b323f","body_text":"The continuously growing amount of information regarding the molecular mechanisms underlying the emergence of tissue abnormalities from benign tumours to invasive malignancies has set the grounds for the use of gene therapy as a rational yet radical therapeutic approach for cancer therapy [ 1 ]. Gene therapy aims to introduce or modify genetic material into target cells, thus altering their function, usually by either restoring a lost function or initiating a new one. Although it was initially employed for the treatment of inherited genetic diseases, gene therapy was soon identified as an effective approach for the treatment of both gynaecological malignancies such as ovarian [ 2 ], cervical [ 3 ], and endometrial cancer [ 4 ] and certain benign gynaecological abnormalities, such as leiomyomas, endometriosis, placental, and embryo implantation disorders [ 5 ]. There are two main strategies for specific and efficient gene delivery to cancer and non-cancer cells, and these involve either viral or non-viral systems, as summarised in  Figure 1  and described below in  Section 2 .\nThe main gene therapy strategies employ mutation compensation, primarily involving the replacement of a mutated tumour suppressor gene or the modification of aberrant gene expression in tumour cells, antiangiogenic, suicide gene therapy, and immunopotentiation approaches [ 5 , 6 ] ( Figure 2 ). This review covers all major advances in gene therapy for gynaecological cancer and benign disorders.\n\nThe main strategies for targeted gene therapy approaches for malignant and benign gynaecological disorders employ both viral and non-viral systems ( Figure 1 ). This section provides a brief overview of the different viral and non-viral gene delivery systems.\nRetroviral vectors: Lentiviral vectors (LVs) are the most widely used retroviruses for therapeutic gene delivery, as they can efficiently transduce both quiescent and dividing cells and thus facilitate safe, efficient, and stable transgene expression [ 7 , 8 ]. They are enveloped, single-stranded RNA viruses with a packaging capacity of ~9 kb [ 7 ] and have already been employed in numerous successful clinical trials for the treatment of various diseases, such as hemoglobinopathies, metabolic and immune disorders, and various cancer types, including gynaecological malignancies [ 9 ]. Moreover, they have been instrumental in the development of chimeric antigen receptor (CAR) T cells, an approach with promising therapeutic perspectives in the field of oncology;\nAdenoviral vectors (Ads): Adenoviral vectors have been extensively used as viral vector platforms primarily due to their broad tropism and the high transduction efficiency of both quiescent and dividing cells, as well as for their capacity to persist as episomal elements within the target cells [ 9 ]. They mainly derive from human serotypes-2 (Ad2) and -5 (Ad5) adenoviruses, which are non-enveloped, double-stranded DNA viruses with an icosahedral capsid able to accommodate up to 45 kb linear, double-stranded DNA [ 10 ]. However, since they are usually associated with potent immune responses [ 11 ] due to pre-existing immunity, extensive research has focused on generating less immunogenic Ad vectors, employing serotypes with low seroprevalence, such as Ad26 and Ad35 [ 12 ]. Furthermore, the generation of conditionally replicative adenoviral vectors (CRAds) by introducing tumour-specific promoters into the Ad genome can lead to higher gene expression [ 13 ]. Regarding gene therapy for malignant and benign gynaecological disorders, Ads have been extensively used as therapeutic vectors in both pre-clinical and clinical studies [ 9 ] for the delivery of vaccines, tumour suppressor genes, suicide genes, and immunomodulatory genes. All of these approaches are discussed in detail below;\nAdeno-associated vectors (AAVs): Adeno-associated vectors are single-stranded DNA viruses with an icosahedral capsid of ~5 kb packaging capacity that depend on adenoviruses to complete their life cycle. The current AAV1 and AAV2 serotypes used are characterised by broad tropism, stable episomal persistence, and reduced immunogenicity, and therefore, they have been widely employed in gene therapy approaches [ 9 ]. Although AAV vectors have not been used yet in clinical gene therapy trials for gynaecological disorders, encouraging data from their pre-clinical assessment presented in this review point toward this direction;\nMeasles virus (MeV): The vaccine strain of MeV is a negative-strand RNA virus with a 16 kb-long genome that presents an attractive oncolytic platform, mainly due to its ability to selectively infect malignant cells, which overexpress the CD46 receptor [ 14 ]. Due to the fact that the monotherapy oncolytic approach using MeV is usually not sufficient to treat advanced-stage malignancies, combination approaches using radiotherapy or chemotherapy, MeV harbouring with therapeutic or immunomodulatory genes, and the use of carrier cells for MeV delivery into tumour cells can lead to successful therapeutic outcomes for a number of malignancies, including gynaecological cancers [ 15 ];\nHerpes Simplex Virus (HSV): HSV-1 and HSV-2 members are enveloped, double-stranded DNA viruses with an icosahedral capsid, which demonstrate attractive vector features, such as a large accommodation capacity, easy production, and high titers [ 16 ]. Apart from their ability to mediate oncolysis upon delivery of the suicide thymidine kinase  (TK)  suicide gene into malignant cells, HSV vectors have also been used to deliver immunomodulatory cytokines into tumour cells and thus elicit a strong anti-tumour immune response. Extensive research has focused on the development of safer HSV vectors, carrying transcriptionally active promoter elements that overcome the latency of the viral genome, aiming to expand HSV pre-clinical and clinical applications [ 16 ].\nThe need for non-viral gene delivery systems arose primarily due to the immunogenicity and cytotoxicity concerns raised by some viral vectors, such as Adenoviruses [ 11 ]. Despite the relatively low transfection efficiency, their use in the field of gene therapy has gained a lot of ground [ 13 , 17 ]. The main non-viral approaches involve the delivery of the desired gene material, e.g., naked or plasmid DNA, either by physical methods, such as electroporation, ballistic DNA, injection, photoporation, magnetofection, sonoporation, and hydroporation [ 13 , 18 ], or chemically, by means of polymeric, lipid-based and inorganic nanoparticles (NPs) [ 13 , 19 ].\nBriefly, the most common types of non-viral physical methods involve the following approaches. Electroporation: The electric pulse creates pores in the cell membrane, allowing the genetic material to enter the target cell. Depending on the target tissue, the electric pulse differs both in strength (high or low) and duration (short or long pulses), with cancer cells usually requiring low field strength (<700 V/cm) with long pulses (milliseconds) [ 13 ]; Ballistic DNA: Bombardment of DNA-coated heavy metal particles into target cells at a certain speed. The precision in DNA delivery makes it a method of choice for ovarian cancer [ 20 ]; Injection: The genetic material is directly injected into tissue by means of a needle. It represents the preferred approach for solid tumours; however, it has a relatively low efficiency [ 20 ]; Photoporation: A laser pulse causes cells to be permeable to DNA, with transfection efficiency being dependent on the focal point and laser frequency [ 13 ]; Magnetofection: It is mostly used for in vitro approaches; a gene-magnetic particle complex is introduced in cell culture, and electromagnets placed below the cell culture generate magnetic fields that mediate its effective sedimentation, thus leading to higher transfection efficiency [ 13 ]; Sonoporation: The generated ultrasound waves cause cell membranes to be permeable to micro-bubbles containing gene products [ 13 ]; Hydroporation: Hydrodynamic pressure generated by the injection of a large volume of DNA in a short period of time increases cell membrane permeability and mediates gene delivery [ 13 ].\nElectroporation: The electric pulse creates pores in the cell membrane, allowing the genetic material to enter the target cell. Depending on the target tissue, the electric pulse differs both in strength (high or low) and duration (short or long pulses), with cancer cells usually requiring low field strength (<700 V/cm) with long pulses (milliseconds) [ 13 ];\nBallistic DNA: Bombardment of DNA-coated heavy metal particles into target cells at a certain speed. The precision in DNA delivery makes it a method of choice for ovarian cancer [ 20 ];\nInjection: The genetic material is directly injected into tissue by means of a needle. It represents the preferred approach for solid tumours; however, it has a relatively low efficiency [ 20 ];\nPhotoporation: A laser pulse causes cells to be permeable to DNA, with transfection efficiency being dependent on the focal point and laser frequency [ 13 ];\nMagnetofection: It is mostly used for in vitro approaches; a gene-magnetic particle complex is introduced in cell culture, and electromagnets placed below the cell culture generate magnetic fields that mediate its effective sedimentation, thus leading to higher transfection efficiency [ 13 ];\nSonoporation: The generated ultrasound waves cause cell membranes to be permeable to micro-bubbles containing gene products [ 13 ];\nHydroporation: Hydrodynamic pressure generated by the injection of a large volume of DNA in a short period of time increases cell membrane permeability and mediates gene delivery [ 13 ].\nNon-viral chemical delivery systems employ particles of different shapes and sizes, known as nanoparticles (NPs), to encapsulate and efficiently transport genetic material, such as DNA and RNA, into target cells via endocytosis. These NPs are divided into the following three classes [ 13 , 19 ].\nPolymeric NPs: These are stable, biodegradable, and water-soluble NPs, which make ideal candidates for drug delivery. They exist as monomers or polymers of different structures, with the most common being nanocapsules and nanospheres. According to shape, they are further divided into polymersomes, dendrimers, and micelles. Polymersomes are artificial vesicles with amphiphilic membranes that display increased cargo capacity and effective cytosol delivery [ 21 ], whereas micelles are quite effective for aqueous drug administration due to their hydrophilic core and hydrophobic coating. Polymeric micelles have already been employed in clinical trials, specifically in a study testing the effective delivery of paclitaxel (PTX), a drug widely used for ovarian cancer treatment [ 22 ]. Dendrimers are more complex, three-dimensional nanocarriers, commonly used for the delivery of nucleic acids and drugs in cancer therapy, usually in combination with poly amidoamine (PAMAM) and polyethylenimine (PEI) polymers [ 23 ]. Despite the toxicity associated with the surface amino groups, PAMAM was the first dendrimer used for gene delivery [ 17 ]; Lipid-based NPs: They are usually spherical vehicles with an internal aqueous compartment and an external lipid bilayer. They are divided into liposomes or lipoplexes, lipid emulsions (or nanogels), and lipid NPs [ 17 ]. Due to their positive charge, cationic liposomes bind to the anionic phosphate group of nucleic acids, forming lipoplexes and interacting with cell membranes, and thus, efficiently deliver nucleic acids into the cell [ 17 ]. Lipid emulsions, which are composed of oil, water, and a surface-active agent, show increased stability and serum resistance, whereas solid lipid NPs can protect nucleic acids from nucleases and are primarily employed for siRNA delivery [ 20 ]; Inorganic NPs: These nanoparticles are composed of inorganic materials, such as iron, gold, and silica, and therefore possess unique electrical, magnetic, and optical properties [ 13 , 19 ]. The most widely used inorganic NPs are gold NPs, primarily due to their low toxicity, while their photothermal properties establish them as good candidates for cancer therapy [ 24 ].\nPolymeric NPs: These are stable, biodegradable, and water-soluble NPs, which make ideal candidates for drug delivery. They exist as monomers or polymers of different structures, with the most common being nanocapsules and nanospheres. According to shape, they are further divided into polymersomes, dendrimers, and micelles. Polymersomes are artificial vesicles with amphiphilic membranes that display increased cargo capacity and effective cytosol delivery [ 21 ], whereas micelles are quite effective for aqueous drug administration due to their hydrophilic core and hydrophobic coating. Polymeric micelles have already been employed in clinical trials, specifically in a study testing the effective delivery of paclitaxel (PTX), a drug widely used for ovarian cancer treatment [ 22 ]. Dendrimers are more complex, three-dimensional nanocarriers, commonly used for the delivery of nucleic acids and drugs in cancer therapy, usually in combination with poly amidoamine (PAMAM) and polyethylenimine (PEI) polymers [ 23 ]. Despite the toxicity associated with the surface amino groups, PAMAM was the first dendrimer used for gene delivery [ 17 ];\nLipid-based NPs: They are usually spherical vehicles with an internal aqueous compartment and an external lipid bilayer. They are divided into liposomes or lipoplexes, lipid emulsions (or nanogels), and lipid NPs [ 17 ]. Due to their positive charge, cationic liposomes bind to the anionic phosphate group of nucleic acids, forming lipoplexes and interacting with cell membranes, and thus, efficiently deliver nucleic acids into the cell [ 17 ]. Lipid emulsions, which are composed of oil, water, and a surface-active agent, show increased stability and serum resistance, whereas solid lipid NPs can protect nucleic acids from nucleases and are primarily employed for siRNA delivery [ 20 ];\nInorganic NPs: These nanoparticles are composed of inorganic materials, such as iron, gold, and silica, and therefore possess unique electrical, magnetic, and optical properties [ 13 , 19 ]. The most widely used inorganic NPs are gold NPs, primarily due to their low toxicity, while their photothermal properties establish them as good candidates for cancer therapy [ 24 ].\n\nAlthough usually considered a single disease, ovarian cancer is a highly heterogeneous group, comprising different histological subtypes with different underlying molecular mechanisms, clinical manifestations, and hence therapeutic approaches [ 25 ]. These include serous, endometrioid, clear-cell, and mucinous carcinoma [ 26 ] epithelial ovarian cancer types, which comprise the majority of ovarian cancers. Gene therapy is an emerging therapeutic approach for ovarian cancer [ 2 ] and is primarily based on viral and non-viral delivery systems ( Figure 3 ) and on strategies such as suicide and antiangiogenic gene therapy, oncolytic virotherapy, mutation compensation, and immunopotentiation ( Table 1 ).\nThe main viral vectors used for the gene therapy of ovarian cancer are summarized in  Table 1  and  Figure 3  and include retroviruses (primarily LVs), Ads, AAVs, HSV, and MeV. Highlighted below are the major advancers for each vector type.\nTargeted gene therapy approaches for ovarian cancer.\nIn a tumour-suppressor cervical-cancer gene therapy approach, Shi et al. employed shRNA to knock-down CD59, a negative regulator of complement activation, and demonstrated enhanced complement-mediated cell damage, increased apoptosis, and the inhibition of tumour growth in vitro and in vivo [ 48 ]. In another study, a lentiviral vector carrying shRNA for Nin one binding protein (NOB1p), a protein frequently overexpressed in ovarian cancer cells, led to the suppression of cancer cell growth and colony formation [ 49 ]. Lastly, Zhang et al. transplanted umbilical cord mesenchymal stem cells (MSCs) engineered to express IL-21 in SKOV3 ovarian cancer xenograft-bearing nude mice, utilising lentiviral vector pUCMSCs-LV-IL-21 and demonstrated a reduction in tumour size through the inhibition of growth and the elevation of IFN-γ and TNF-α [ 47 ].\nThese viral systems have been widely used in ovarian cancer gene therapy employed in suicide gene therapy, antiangiogenic, and oncolytic virotherapy, alone or together with immunopotentiation, as well as mutation compensation approaches. With regards to suicide gene therapy, Rawlinson et al. [ 27 ] developed an adenoviral vector carrying the  HSV-TK  gene under the tumour-specific promoter HE4 and demonstrated increased cell death by prodrug ganciclovir (GCV) treatment in vitro, in ovarian cancer cell lines. When Zhou et al. applied the above system in vivo in a murine ovarian cancer model, they observed an enhanced anti-tumour effect and increased apoptosis, as well as a reduction in micro-vessel density [ 28 ]. In a similar approach, but using the NTR-CB1954 system instead, White et al. performed an adenoviral delivery of both  NTR  and CB1954 in ovarian cancer cells and achieved increased survival in both in vitro and in vivo [ 29 ].\nThe antiangiogenic approach using Ads and AAVs has also yielded promising results in ovarian cancer, as demonstrated by several studies described in this section. Specifically, Tuppurainen et al. [ 32 ] used a combined strategy using adenoviral vectors expressing important genes in angiogenesis, such as vascular endothelial growth factor  (VEGF) 2  receptor and angiopoietin growth factor receptor  Tie2 , with or without chemotherapy, and demonstrated reduced tumour growth and ascites formation. In a different approach, Hampl et al. [ 30 ] constructed the adenoviral vector, Ad-hEndo, and used it to mediate the transfer of human endostatin ( COL18A1 ) in ovarian cancer ascites in mice, demonstrating a significant down-regulation in ascites formation, reduced tumour growth and vascularity, as well as prolonged animal survival [ 30 ]. In order to enhance transfection efficiency and escape antibody neutralisation in vivo, Yang et al. encapsulated the above vector into PEG-PE cationic liposomes and demonstrated the suppression of ovarian cancer growth in vivo, in ovarian cancer mouse models [ 31 ]. Similarly, the adeno-associated-mediated delivery of  P125A- COL18A1 , a mutated endostatin gene, led to the inhibition of tumour growth and blood vessel formation [ 33 ], while the administration of Kringle 5 (K5) of human plasminogen, a potent angiogenesis inhibitor, by means of a recombinant AAV vector led to the inhibition of VEGF and tumour angiogenesis [ 34 ]. Lastly, the use of AAVrh10.BevMab, a rhesus serotype 10 adeno-associated vector coding for bevacizumab, demonstrated a significant reduction in tumour growth, reduced angiogenesis, and increased survival in mice with ovarian cancer [ 35 ].\nRegarding the oncolytic virotherapy strategy, apart from HSV, the Vesicular Stomatitis Virus (VSV), and MeV described below, adenoviruses have also been employed and led to significant anti-tumour effects. Specifically, Han et al. designed the novel oncolytic Ad-M4 targeting signal transducer and activator of transcription 3 ( STAT3)  in ovarian cancer and showed that the aberrant expression and constitutive activation of STAT3 were key factors in cisplatin resistance in both ovarian cancer cell lines and ovarian cancer tissues samples [ 43 ]. The authors showed significant and selective inhibition of ovarian cancer cell survival and cisplatin-induced apoptosis both in vitro and in vivo following Ad-M4 administration [ 43 ]. In a similar approach, Wang et al. constructed the manganese superoxide dismutase-armed oncolytic Ad ZD55-MnSOD and used it together with cisplatin, achieving enhanced cisplatin-induced growth suppression and apoptosis in ovarian cancer cells lines and ovarian tumour xenografts [ 44 ].\nFurthermore, and regarding the mutational compensation approach, most studies utilising Ads target either the mutant p53, a protein with a variety of anti-cancer functions in gynaecological cancers [ 78 ], or key apoptosis pathway regulators, such as Bim and Bid. In the former approach, which made its way to Phase II/III clinical trials, the replacement of the altered  p53  gene leads to significant anti-tumour effects [ 37 , 38 ]. Specifically, the in vitro and in vivo application of Ad-p53 resulted in the activation of apoptosis and taxol-sensitisation in ovarian cancer cells [ 38 ], while Gendicine, a gene therapy product approved for clinical use by the Chinese FDA in 2003, is now demonstrating very promising results, leading to a 90% patient response rate and 100% peritoneal effusion resolution [ 37 ]. In a recent study, Dai et al. [ 36 ] constructed two tumour-specific oncolytic adenoviruses coding for  Bim  (Ad-Bim) or truncated  Bid  (Ad-tBid) and performed gain-of-function assays in nine ovarian cancer cell lines. The authors observed increased cisplatin sensitisation and apoptosis following treatment with Ad-tBid in vitro and ex vivo, while following Ad-tBid intraperitoneal administration in mice with peritoneal disseminated ovarian cancer, they observed enhanced cisplatin-mediated anti-tumour effects [ 36 ]. Lastly, in regard to the immunopotentiation approach, HER2/neu, a protein shown to be overexpressed in ovarian cancer [ 79 ], was delivered to dendritic cells (DCs) by an AAV and primed these cells to express a self-tumour antigen. The above DC-loading elicited a strong MHC class I antigen presentation and led to a potent cytotoxic T lymphocyte response against ovarian cancer cells [ 46 ]. However, in view of the fact that the frequency of  HER2  overexpression in ovarian cancer is low [ 80 ], the above approach may have limited applicability.\nThey comprise the main viral systems utilised in oncolytic virotherapy/suicide gene therapy approaches for ovarian cancer gene therapy. More specifically, Hartkopf et al. [ 45 ] designed a recombinant strain of MeV that carried a fusion cytosine deaminase (CD) protein, which enhanced the chemosensitivity of ovarian cancer cells in 5-fluoroucacil (5-FU) and used it in vitro and in vivo investigations. The authors demonstrated the effective infection and lysis of both human ovarian cancer cell lines and primary tumours [ 45 ]. In another study, Hanauer et al. [ 39 ], using a bispecific oncolytic MeV, with both HER2 and EpCAM as target receptors, demonstrated a significantly enhanced lysis of tumour cells in xenograft mice, highlighting the superiority of bispecific against monospecific oncolytic viruses [ 39 ]. Regarding HSV viruses, Goshima et al. [ 41 ] and Thomas et al. [ 40 ], in two independent studies, employed HSV in combination with immunostimulatory cytokines, such as GM-CSF and IL-12, respectively. The intraperitoneal injection (IP) of HSV-HF10 in a murine ovarian cancer murine model, together with the anti-tumour effect of GM-CSF, led to a reduction in tumour size and an increase in anti-tumour immune response [ 41 ]. Similarly, the intraperitoneal administration of HSV engineered to express IL-12 led to cytotoxic effects, reduced metastasis, and increased CD8 +  T cell response [ 40 ]. Lastly, the use of VSV-GP, a potent oncolytic virus pseudotyped with the lymphocytic choriomeningitis virus (LCMV) envelope glycoprotein, together with the Jak1 and Jak2 inhibitor Ruxolitinib, led to transient tumour remission in mice following IP administration [ 42 ].\nStrategies of this category utilise DNA/RNA plasmids to alter gene expression via transient transfection or nanoparticles to target a molecule expressed on the surface of cancer cells ( Figure 3 ,  Table 1 ).\nThe use of these approaches has yielded therapeutic effects through suicide gene therapy, immunopotentiation, and mutation compensation strategies. More specifically, Sher et al. [ 50 ] constructed the hEndoyCD fusion protein, composed of the antiangiogenic endostatin and the  Escherichia coli  cytosine deaminase (CD) domain which converts 5-fluorocytosine (5-FC) to 5-FU, and delivered it to ovarian cancer cells via a plasmid vector (SV-hEndoyCD). The authors reported tumour-specific growth inhibition and increased survival in xenograft mouse models [ 50 ]. Regarding immunopotentiation approaches, two reports provide evidence of survival improvement [ 57 , 58 ]. Specifically, Hu et al. employed human umbilical cord CD34 +  stem cells transfected with the pIRES2-IL-21-EGFP plasmid, carrying the anti-tumour cytokine IL-21, and demonstrated a therapeutic effect in ovarian cancer xenografts, possibly due to tumour-specific NK cytotoxicity, as a result of elevated levels of IFN-γ and TNF-α. Despite the gradual decrease in IL-21 in mice tumour tissues, overall extended survival was observed in these mice [ 57 ]. Similarly, Fewell et al. employed a pmIL-12/PPC vehicle carrying the anti-cancer IL-12 cytokine and demonstrated an efficient treatment for disseminated ovarian cancer as a result of the significant VEGF decrease and IL-12 and IFN-γ increase following IP administration [ 58 ].\nRegarding mutation compensation, a plethora of studies have demonstrated significant anti-tumour effects following this approach [ 60 , 61 , 62 , 63 , 64 , 65 , 67 , 68 , 69 , 70 ]. Specifically, Wu et al., demonstrated a reversal of the malignant phenotype in vitro following the transfection of A2780 epithelial ovarian cancer (EOC) cells with a plasmid carrying the wild-type phosphatase and tensin homolog ( PTEN)  gene [ 60 ]. The above cells showed apoptosis induction, G1 arrest, and a decrease in migration and invasion [ 60 ]. In a different study, transfection with pcDNA3.1-p16/pcDNA3.1-vi5-hisA-eEF1A2 plasmids led to eEF1A2 down-regulation through the increase in tumour-suppressor  p16  and in turn inhibited cancer cell growth [ 61 ]. In a similar approach, Lu et al. [ 62 ] demonstrated a significant suppression of ovarian cell proliferation and migration in vitro and in vivo by the up-regulation of p16. The tumour-suppressing effects of the WW domain containing the oxidoreductase  (WWOX)  gene have been employed in targeted mutation compensation approaches in ovarian cancer, leading to significant results. Yan et al. [ 63 ] used a recombinant plasmid to transfect ovarian cancer stem cells and showed decreased proliferation, cell cycle arrest during the G0/G1 phase, and the induction of apoptosis through the up-regulation of caspase-3. A similar result was observed by Xiong et al. [ 65 ], employing WWOX delivery in liposomes, as described below. Recently Zhao et al. [ 64 ] showed that  WWOX  activation might inhibit chemoresistance and cell death induction in PTX-treated EOC cells. Specifically, the overexpression of WWOX in PTX-sensitive EOC cells using the pcDNA3.1-WWOX plasmid demonstrated the induction of apoptosis and the inhibition of autophagy and cell proliferation following PTX treatment [ 64 ]. The importance of the down-regulation of metastasis-associated in colon cancer 1  (MAAC1),  metastasis-associated 1  (MTA1),  and cyclooxygenase 2  (COC2)  genes in ovarian cancer induction and progression was demonstrated by several studies, employing the down-regulation of their expression by shRNA/siRNA strategies [ 67 , 68 , 69 , 70 ]. Specifically, Zhang et al. used an shRNA plasmid to knock-down  MACC1  in ovarian carcinoma (OVCAR)-3 cells to demonstrate the inhibition of proliferation, migration, invasion, and induction of apoptosis both in vitro and in vivo in mouse models [ 67 ], while the siRNA interference strategy to knock-down  MTA1  in ovarian cancer cell line A2780 resulted in reduced migration, invasion, and adhesion and increased apoptosis in vitro [ 68 ]. Finally, the knock-down of cyclooxygenase 2 ( COX2 ) by two independent groups [ 69 , 70 ] highlighted the role of the  COX2  gene in ovarian cancer cell proliferation, invasion, migration, and growth.\nThe use of these systems, primarily in suicide gene therapy and mutation compensation targeted approaches against ovarian cancer, has also yielded significant results. Specifically, Bai et al. transferred gelonin toxin in ovarian cancer cells using cationic heparin PEI (HPEI) nanogels and managed to reduce cancer cell growth and induce apoptosis [ 51 ]. Furthermore, Huang et al. [ 52 ] delivered diphtheria toxin subunit-A (DT-A) DNA in mice with ovarian tumours by means of cationic polymer administered IP, placing it under the control of a human epididymis protein 4 (HE4) promoter, whose activity is increased in ovarian cancer cells. The latter promoter was also used to drive the  TK  gene, leading to the inhibition of tumour growth and increased survival in mice upon delivery [ 52 ]. In a similar approach, Cocco et al. [ 53 ] employed the IP administration of NPs engineered to deliver a plasmid encoding for DT-A and succeeded in inhibiting ovarian tumour growth. Recently, Sun et al. [ 54 ] managed to successfully deliver a GCV-converted HSV- tTK  in ovarian cancer cells employing Tat/pDNA/C16TAB (T-P-C) NPs, demonstrating targeted delivery in mouse models with human ovarian cancer.\nRegarding the mutation compensation approach using nanoparticles, this involves the delivery of siRNA in NPs, targeting  EGFR  [ 66 ], the receptor of  Clostridium perfringens  enterotoxin claudin-3  (CLDN3 ) [ 74 , 75 ] and hypoxia-inducible factor 1-alpha  (HIF1-a ) [ 76 ]. To this end, Dickerson et al. knocked down  EGFR , an oncogene highly expressed in ovarian cancer cells and implicated in cell proliferation, migration, and differentiation, using siRNA delivered in nanogels and showed increased sensitivity of these cells to docetaxel [ 66 ]. Similarly, the knock-down of  CLDN3  by siRNA [ 74 ] or shRNA [ 75 ] led to the inhibition of tumour growth, a reduction in cell proliferation and angiogenesis, and the induction of apoptosis. Lastly, the silencing of  HIF1-a , often overexpressed in ovarian cancer by means of siRNA in NPs, led to the effective inhibition of cell proliferation [ 76 ]. Regarding lipoplexes, dendrimers, and liposomes, a significant number of studies [ 55 , 56 , 59 , 71 , 72 , 73 , 77 , 81 ] have reported efficient delivery in ovarian cancer cells, highlighting the significant contribution of NPs in targeted gene therapy applications. Concerning the suicide gene therapy approach, He et al. [ 55 ] used FRα-targeted folate-modified lipoplex loaded with an hTERT promoter-regulated plasmid encoding VSV matrix protein (MP) (F-LP/pMP) and transfected SKOV3 ovarian cancer cells. The F-LP/pMP treatment resulted in the significant inhibition of tumour growth, as demonstrated by the inhibition of cell proliferation, apoptosis induction, and the suppression of tumour angiogenesis, as well as by the extended survival of mice in a SKOV-3 tumour model [ 55 ]. Moreover, Kobayashi et al. [ 56 ], employing a PAMAM dendrimer to deliver carbonyl reductase 1  (CRB1 ), which reduces micro-vessel density and leads to apoptosis, observed increased survival in mice and the decreased proliferation and dissemination of cancer cells.\nFurthermore, in a study by Yang et al. [ 59 ], the delivery of the human pro-apoptotic gene  hPNAS-4  into ovarian cancer cells using cationic liposomes led to the efficient inhibition of tumour growth, the induction of apoptosis, and the inhibition of angiogenesis both in vitro and in vivo. Furthermore, the inhibition of the Wilms tumour 1  (WT1 ) gene, overexpressed in ovarian carcinoma but not in normal tissue, by an antisense oligodeoxynucleotide (ASODN) delivered in liposomes, led to the inhibition of ovarian cancer cell proliferation, cell cycle arrest, and increased apoptosis in vitro [ 71 ]. The delivery of shRNA against STAT3 in liposomes resulted in a reduction in the growth of intraperitoneal ovarian cancer through the inhibition of cell proliferation and apoptosis induction [ 72 ], while the delivery of siRNA against NOTCH1 in cationic cholesterol derivative-based liposomes led to the inhibition of growth and apoptosis induction in vitro [ 73 ]. As mentioned above, Xiong et al. [ 65 ] delivered WWOX in ovarian cancer cells by means of liposomes and documented decreased cancer cell growth and apoptosis induction. Lastly, the utilisation of the CRISPR/Cas9 system in a folate receptor-targeted cationic liposome (F-LP) to target the DNA methyltransferase 1  (gDMNMT1)  gene, whose overexpression results in inactivation or tumour suppressor genes and increased ovarian cancer cell resistance, resulted in the inhibition of ovarian cancer tumours both in vitro and in vivo [ 77 ].\nLarge-scale microarray analysis has highlighted the role of many microRNAs [miR(s)] in different types of cancer, including ovarian and cervical cancer [ 82 ]. miRNAs in ovarian cancer have shown to have either a tumour-promoting or tumour-suppressing role, depending on whether their expression is up-regulated or down-regulated, respectively, and hence can be employed both as therapeutic targets, as well as biomarkers for diagnosis and/or prognosis [ 83 ]. Regarding ovarian cancer, Iorio et al. [ 84 ] performed an initial and comprehensive miRNAs comparison between normal and ovarian cancer tissues and showed that miR-14, mir-199a, miR-200a, miR-200b, and miR-200c were up-regulated in ovarian cancer and thus acted as oncogenic miRs (oncomiRs), while miR-15, miR-16, mir-140, mir-145, mir-199a, and miR-125b1 were down-regulated, pointing toward a tumour-suppressing role (tumour suppressor miRs). Further studies highlighted novel players, such as miR-214, miR-150, miR-140-5P, miR-21, miR-29a, and let-7a and provided more insights into the role of different miRNAs in ovarian cancer.\nMore specifically, Yang et al. [ 85 ] showed that miR-214 and miR-150 are highly up-regulated in ovarian cancer and act as oncomiRs, inhibiting the expression of  PTEN  tumour suppressor gene, thus initiating the protein kinase B (ACT) pathway. On the other hand, the prevention of miR-PTEN interaction can inhibit ovarian cancer, primarily through the inhibition of cancer cell proliferation and chemotherapeutic drug resistance. Therefore, the down-regulation of both miR-214 and miR-150 expression could potentially have a therapeutic effect. Similarly, experiments by Lan et al. [ 86 ] showed that miR-140-5p expression is significantly decreased in ovarian cancer patients, while when up-regulated, it leads to the induction of ovarian cancer cell proliferation and migration. The authors also observed that miR-140-5p down-regulation could reduce sensitivity to cisplatin, and therefore, the up-regulation of the specific miRNA could lead to both the inhibition of proliferation and migration and an enhancement of cisplatin sensitivity. Regarding miR-21, Echevarria-Vargas et al. [ 87 ] showed that miR-21 is also associated with cisplatin resistance among ovarian cancer cells. Its expression was found to be up-regulated in cisplatin-resistant ovarian cancer cells, and therefore targeting miR-21 can lead to reduced tumour characteristics in the specific cells. The down-regulation of miR-29a in ovarian cancer tissues is associated with apoptosis [ 88 ] while restoring let-7A-3/let-b expression leads to a reduction in tumour growth both in vitro and in vivo and the inhibition of invasion and metastasis [ 89 ]. Furthermore, increased let-7a is associated with increased survival rates in patients receiving platinum-based chemotherapy [ 90 ]. Similarly, the up-regulation of miR-200 family members is associated with decreased metastasis, primarily through reduced epithelial–mesenchymal transition (EMT) and tumour angiogenesis. Pecot et al. [ 91 ] reported that the miR-200 family inhibits blood vessel formation by targeting IL-8 and CXCL-1; therefore, the expression of the miR-200 family to the tumour epithelium results in decreased tumour cell metastasis and angiogenesis. Moreover, the observation that miR-183 and miR-22 expression was elevated in the low, compared to the high form of metastatic cancer, coupled with the fact that increasing the expression of miR-183 and miR-22 resulted in EZRIN reduction, led to the conclusion that EZRIN inactivation through miR-183 and miR-22 can inhibit ovarian cancer metastasis [ 92 ]. Lastly, Aqeilan et al. [ 93 ] demonstrated that miR-15 and miR-16 are involved in the multi-drug resistance of ovarian cancer cells by acting on Bcl-2, while Luo et al. [ 94 ] reported the tumour suppressing role of miR-126 in cancer SKOV3 cells, by demonstrating a miR-126-mediated P21 (RAC1) activated kinase 4 (PAK4) down-regulation.\nCervical cancer remains the fourth most common and most lethal cancer type in women, despite regular screening and prevention strategies [ 95 ]. Targeted gene therapy presents a promising approach for the treatment of the specific malignancy and focuses primarily on mutation compensation strategies, suicide gene therapy, oncolytic virotherapy, antiangiogenic strategies, immunopotentiation, and drug resistance therapies [ 3 ], employing both viral and non-viral systems.\nThe main viral vectors used for the gene therapy of cervical cancer are LVs, Ads, and AAVs ( Figure 4 ,  Table 2 ).\nCervical cancer gene therapy approaches employ LVs primarily for mutation compensation, antiangiogenic, and suicide gene therapy strategies. The restoration of important tumour suppressor genes’ expression, such as tyrosine phosphatase receptor J  (PTPRJ),  asparaginase and isoaspartyl peptidase 1  (ASRGL1),  and homeobox-containing 1  (HMBOX1) , has been successfully achieved with LVs. Specifically, Yan et al. [ 96 ] used pSicoR-PTPRJ LV to overexpress  PTPRJ,  a tumour suppressor gene whose expression is down-regulated in human cervical cancer tissues and demonstrated significant suppression of cell viability, migration, and growth in HPV-negative C33A cells. On the contrary, the knock-down of PTPRJ expression in the above cervical cancer cell line led to increased resistance to 5-FU-mediated apoptosis, verifying the importance of elevated PTPRJ expression for cervical cancer prevention. Moreover, Zhou et al. used a lentiviral shRNA to knock-down HMBOX1 expression in HeLa and C33A cancer cells and demonstrated increased radiosensitivity as a result of telomere shortening [ 97 ]. Lastly, the knock-down of ASRGL1 expression in SiHa cells by means of shRNA led to decreased proliferation, possibly through the reduced expression of CDK2 and cyclin A2 and the induction of apoptosis [ 98 ], which was characterised by the increased expression of the pro-apoptotic Bax and the decreased expression of anti-apoptotic Bcl-2. Regarding the antiangiogenic approach, Qi et al. used a lentiviral shRNA-VEGF construct to knock down VEGF expression in vitro and in vivo in nude mice and was able to inhibit tumour growth and tumour radiosensitivity [ 99 ]. One of the most promising approaches for cervical cancer gene therapy using lentiviruses is the genetic modification of T-cell receptors (TCRs) to target tumour-specific antigens. Based on the above, Jin et al. developed E7-specific T cells and achieved regression of HPV-positive mouse tumours [ 100 ]. The specific approach is currently under clinical trial  NCT02379520  and is being used on patients with metastatic cervical cancer [ 101 ]. Regarding suicide gene therapy, the LV-mediated targeting of IAA and CD5 has been employed, while the GlNaTK retroviral vector was used to transfer the suicide  TK  gene into tumour cells. More specifically, Zhang et al. combined SEC3-ES LV targeting IAA and cisplatin and showed cell cycle arrest at the S phase, as well as the induction of apoptosis [ 102 ]. Similarly, Hao et al. studied the synergistic effect of two suicide gene systems, such as  CD/5-FC  and  HSV-TK/ganciclovir  (GCV), and demonstrated the inhibition of proliferation and enhanced apoptosis in vitro following lentiviral delivery combined with ultrasonic radiation [ 103 ]. Lastly, Chen et al. employed the GlNaTK retroviral vector to mediate  TK  transfer, which led to the inhibition of cell growth, increased apoptosis, and tumour radiosensitisation to cobalt-60 following GCV administration [ 104 ].\nTargeted gene therapy approaches for cervical cancer.\nGene therapy strategies for cervical cancer. They employ viral vectors, primarily adenoviral and adeno-associated viral vectors and lentiviral vectors, and non-viral systems, such as DNA plasmids, nanoparticles, CRISPR/CaS9, and shRNA/siRNA/miRNA approaches.\nThey comprise the majority of viral vectors used for cervical cancer gene therapy, often utilised for tumour suppressor gene restoration or blocking of oncogenic expression. The  p53  gene, a key regulator of cell proliferation, apoptosis, and genetic stability [ 141 ], plays a fundamental role in most gynaecological cancers and has therefore been a major candidate for most targeted gene therapy approaches [ 78 ]. More specifically, since the  p53  gene is often inactivated by HPV E6 protein in most cervical cancers [ 142 ]; either its delivery or inhibition of the E6 protein could result in significant antitumour effects. Indeed, Su et al. employing Genidicine ® , a gene therapy product approved in 2003 by the China Food and Drug Administration (CFDA) for head and neck cancer gene therapy [ 143 ], showed that the injection of a recombinant human adenovirus engineered to express wild-type  p53  gene (rArd-p53) in cervical cancer patients, may lead to an increased 5-year overall survival rate [ 105 ]. Moreover, Kajitani et al., using siRNA for E6 protein in HeLa cells, demonstrated successful  p53  transduction following adenovirus treatment [ 106 ]. When the above transduction in HeLa cells was combined with PTX, it led to enhanced growth inhibition and apoptosis, as demonstrated by Liu et al. [ 38 ]. A further approach in tumour repressor gene restoration strategy is the use of rAd-ER-DN, an adenovirus expressing a dominant-negative estrogen receptor gene  (DNER),  to regulate estrogen imbalance, often associated with cervical cancer development. The use of rAd-ER-DN alone [ 107 ] or in combination with cisplatin and PTX [ 108 ] led to the blocking of cell proliferation and the induction of apoptosis in Ca Ski cervical cancer cells, primarily through the decrease in E6 and E7 oncoprotein expression. Regarding the oncolytic virotherapy approach, the use of five different Ads, i.e., rAd-M5, rAd-M6, rAd-53 [ 112 ], rAd-VEGFI-251, and rAd-H101 [ 113 ], demonstrated significant anti-tumour effects both in vitro and in vivo and led to increased apoptosis of cancer cells. rAd-M5 expressed the HPV  E2  gene, which negatively regulated E6 and E7 expression and led to growth suspension and increased apoptosis [ 110 ], while rAd-M6 employing antisense HPV16 E6/E7 DNA inhibited the expression of E6 and E7 and induced apoptosis in vitro and in vivo [ 111 ]. In another study, Xiao et al. used an oncolytic adenovirus expressing the VEGF inhibitor VEGI-251 and demonstrated the induction of apoptosis and the inhibition of neovascularisation [ 114 ], while Wang et al., employing rAd-p53 carrying the promoter of early growth response (Egr-1), and the anti-apoptotic TNF-related apoptosis-inducing ligand ( TRAIL)  gene showed increased radiotherapy-induced apoptosis in cell lines and tumour size reduction in xenograft mice [ 112 ]. Recently, Duan et al., using rAd-H101 oncolytic adenovirus targeting the E6 gene demonstrated enhanced radiation anti-tumour activity, primarily in C33A cells [ 113 ]. As far as the suicide gene therapy approach is concerned, the use of an Ad carrying the human lactoferrin (hLf), a glycoprotein involved in solid tumour growth reduction, demonstrated the inhibition of tumour growth, due to cell cycle arrest in the G2/M phase in HeLa cells and a prolonged survival in mice with cervical cancer [ 109 ]. Lastly, immunopotentiation strategies in cervical cancer gene therapy show promising results and involve both adenoviruses and adeno-associated viruses. In particular, the delivery of hIL-12, a cytokine with potent anti-cancer activity via the adenoviral vector (rAd-hIL12), led to the inhibition of tumour growth and an increase in survival rates in murine models [ 115 ]. Similarly, the delivery of the anti-tumour hIL-15 cytokine using the rAAV2-hIL15 led to the inhibition of growth in HeLa cells in vitro and in vivo in murine models [ 116 ], while rAAV2-LIGHT, an AAV vector carrying  LIGHT , a TNF ligand, succeeded in suppressing tumorigenesis in a murine cervical cancer model [ 117 ].\nStrategies of this category utilise primarily NPs to target a molecule expressed on the surface of cancer cells, as well as DNA and RNA plasmids or miRNAs to alter gene expression via transient transfection ( Figure 4 ,  Table 2 ).\nGene therapy strategies by means of nanoparticles focus primarily on mutation compensation and immunopotentiation strategies. Regarding the former approach, Liu et al. [ 144 ] developed polyethylene glycol-polylactic acid (PEG-PLA) NPs linked to folate and targeted cancer cells through the folate receptor α (FRα), a membrane-bound protein mediating folate uptake. Although the authors observed enhanced gene transfection efficiency, higher compared to naked DNA, and reduced cytotoxicity, the strategies targeting FRα receptors are hampered by its heterogeneous expression among cervical cancer patients [ 145 ]. A few years later, Yang et al. overexpressed FRα, previously shown to be highly expressed in cervical cancer, using an FRα-targeted liposome (FLP) to deliver a pigment epithelium-derived factor  (PEDF)  gene into HeLa cells and observed significant anti-tumour activity, as demonstrated by significant growth inhibition, the suppression of adhesion and invasion, and cancer cell migration in vitro [ 130 ]. Lechanteur et al. used PEG-lipoplexes to mediate siRNA delivery against E6 and E7 oncoproteins and the anti-apoptotic myeloid cell leukaemia 1 ( MCL1)  gene; this was followed by the successful and prolonged delivery of active siRNAs by PEGylated lipoplexes, suggesting that this approach can treat HPV-induced lesions in vitro and in vivo in murine models [ 125 ]. Moreover, Han et al. using an AP-PAMAM derivative, induced an anti-proliferative effect in vitro by suppressing cancer cell migration and invasion, as well as the induction of cell-cycle arrest at the S phase [ 126 ]. Recently, the use of CRISPR/Cas9 against the HPV16  E6  and  E7  encoding genes delivered in nanoliposomes led to a significant reduction in tumour size in murine mouse models [ 120 ]. Nanoparticles, composed of non-viral vector PBAE546 and a CRISPR/Cas9 recombinant plasmid against HPV16 E7 protein, led to the inhibition of growth in xenograft tumours and the reversal of the cervical epithelial malignant phenotype of the HPV16 transgenic mice [ 123 ]. Finally, regarding the immunopotentiation approach, Wang et al. [ 135 ] and Yang et al. [ 136 ] employed a liposome-encapsulated plasmid encoding  CXCL10 , an important cytokine with a potential therapeutic role in cervical cancer; they demonstrated an inhibition of growth and angiogenesis, which was associated with induction of apoptosis, mediated by the modulation of E6 and E7 oncoproteins expression both in vitro and in vivo.\nThe use of different plasmids in cervical cancer gene therapy approaches focuses on mutational compensation, antiangiogenesis, immunopotentiation and chemoresistance. With regards to blocking oncogenic expression, Hu et al. employed the genome editing approach using both the CRISPR/Cas9 and transcription-activator-like nucleases  (TALEN)  systems to disrupt the HPV16  E7  oncoprotein coding gene [ 122 ]. The data documented induction of apoptosis and growth inhibition in HPV16-positive human cancer cells. The latter approach is being studied in Phase I clinical trials for the treatment of CIN-1 patients with HPV16 and HPV18 infection [ 101 ]. Recently, two groups demonstrated the therapeutic effect of targeting HPV  E6  and  E7  genes. Ling et al., using the CRISPR/Cas9 approach to delete the HPV18  E6  and  E7  genes, achieved robust knock-out of these proteins and increased apoptosis and tumour size reduction [ 119 ]. In an attempt to compare the efficiency of CRISPR/Cas9 with the established TALEN approach, Gao et al. employed the former system against the HPV16  E7  gene and succeeded in reverting cervical carcinogenesis, both in vitro and in vivo [ 124 ]. Another tumour suppressor gene with clinical importance in cervical cancer gene therapy is the retinoblastoma protein zinc finger gene 1  (RIZ1)  since it can induce apoptosis and cell cycle arrest. Cheng et al. demonstrated that overexpression of RIZ1 expression in HPV16-positive cervical cancer cells could lead to impaired cell proliferation and increased apoptosis [ 131 ]. When this overexpression is combined with radiotherapy, it leads to increased apoptosis and DNA damage in HeLa and SiHa cells [ 146 ]. Blocking oncogenic expression can also be mediated by plasmids, primarily via the siRNA and shRNA approaches. Important genes such as X-linked inhibitor of apoptosis ( XIAP),  matrix metalloproteinase  (MMP),  and telomerase reverse transcriptase  (hTERT)  were targeted using the above approaches with promising results. Specifically, Wang et al. employed an RNA interference approach to silence the anti-apoptotic  XIAP  and demonstrated cleaved caspase-3 activation and apoptosis in tumour tissues [ 127 ]. Similarly, targeting of  hTERT , which is often upregulated in cervical cancer and implicated in tumour cell invasion and proliferation, resulted in induction of growth inhibition, along with enhanced radiosensitivity in vitro [ 128 ], reduced cell proliferation, migration and invasion of cancer cells in vitro, as well as in inhibition of cell growth and tumorigenesis in nude mice [ 129 ]. The direct knock-down of the MMP9 protein involved in metastasis [ 132 ] or the indirect targeting of PTX3 [ 133 ], a molecule necessary for MMP2 and MMP-9 expression using shRNA plasmids, led to a reduction in cell migration and invasion, both in vitro and in vivo. In addition to the use of siRNAs for mutational compensation, their use in antiangiogenic strategies has also been demonstrated. Specifically, Zhang et al. used RNA interference to knock-down  Net1 , shown to be overexpressed in cervical cancer tissues, and demonstrated decreased proliferation, migration, and angiogenesis in SiHa cells, as well as a reduction in angiogenesis and tumour growth in vivo, via VEGF down-regulation [ 134 ]. Regarding immunopotentiation, E6 and E7 oncoproteins have been the optimal targets for immunotherapy and DNA vaccine production. Sun et al. developed a DNA vaccine using a plasmid encoding calreticulin (CRT) and E7 protein and managed to evoke a potent CD8 T-cell response, which could ultimately lead to anti-tumour effects and survival [ 137 ]. The co-administration of DNA encoding IL-2 in tumour-bearing mice elicited a stronger CTL response and enhanced anti-tumour effects [ 138 ]. The above approach is under clinical trial  NCT00988559  [ 101 ]. Lastly, the issue of chemoresistance is also addressed by gene therapy to induce sensitivity in cervical cancer cells to chemotherapy. Examples of such approaches with promising results include the up-regulation of hTERTC27 and BRACA1-interacting protein (BRIP1) expression. More specifically, Lin et al. demonstrated that the overexpression of hTERTC27 expression led to increased sensitivity in HeLa cells to 5-FU, followed by inhibition of cell proliferation and increased apoptosis, primarily through the down-regulation of anti-apoptotic Bcl-2 expression and the up-regulation of pro-apoptotic caspases 3 and 9 [ 69 ]. Similarly, Liu et al. found that the overexpression of  BRP1  leads to enhanced cisplatin-mediated anti-tumour activity, increased apoptosis, and suppression of angiogenesis in vitro [ 140 ].\nmiRNAs involved in cervical cancer induction and development are classified into oncogenic miRNAs (oncomiRs) and tumour suppressor miRNAs (tumour suppressor miRs) [ 147 , 148 ] ( Figure 4 ), and therefore, their down-regulation or overexpression, respectively, can be curative, while their presence in patients’ serum is considered as an important tool for cancer diagnosis and prognosis. More specifically and regarding oncomiRs, the most important ones include miR-10a, miR-19, miR-20, miR-21, miR-133b, and miR-886-5p [ 147 ]. Long et al. showed that miR-10a suppresses cell adhesion molecule L1, such as  (CHL1),  and thus leads to enhancement of tumour growth, metastasis, and invasion [ 149 ], while miR-19 is overexpressed in cervical cancer, and its silencing in SiHa cells led to a reduction in the proliferation and induction of apoptosis, through the up-regulation of Bax and down-regulation of Bcl-2 expression [ 150 ]. MiR-20 is a positive regulator of tyrosine kinase non-receptor 2  (TNKS2),  an oncogene involved in metastasis and invasion [ 151 ]. In contrast, miR-21, which is the most well-known oncogenic miRNA, acts as a negative regulator of the tumour suppression gene programmed cell death 4  (PDCD4) , which normally inhibits cell proliferation and induces apoptosis [ 152 ], thus promoting tumour growth. Silencing by means of siRNA in cervical cancer cell lines led to inhibition of cell proliferation and induction of cell death by autophagy and caspase 3/7-mediated apoptosis [ 153 ]. Moreover, through targeting and regulating  CCL20 , a gene involved in tumour differentiation and metastasis, miR-20 was implicated in cervical squamous carcinogenesis [ 152 ]. Lastly, miR-133b is implicated in malignant transformation, originating from ERK and AKT1 signalling pathways [ 154 ], while miR-886-5p is negatively regulating the expression of the apoptotic protein Bax, as its silencing in SiHa cells led to induction of Bax expression and sequentially to apoptosis [ 155 ]. Protective miRNAs with important tumour suppressor activity primarily include miR-143, miR-148b, miR-211, miR-187, miR-641, miR-664, miR-let-7a, and miR-129-5p. More specifically, Liu et al. showed that the overexpression of miR-143 in HeLa cells led to suppression of Bcl-2 expression and thus the inhibition of cell proliferation and the induction of apoptosis [ 156 ]. Similarly, the transfection of miR-187 in SiHa cells also led to the downregulation of Bcl-2 and, in turn, to a reduction in cell growth and increased apoptosis [ 157 ]. Mou et al. employing miR-148b overexpression in HeLa cells demonstrated induction of G1/S cell-cycle arrest and apoptosis, mediated by caspase-3 [ 158 ], while Liu et al. showed that the down-regulation of IAP expression in SiHa cells following miR-211 overexpression, led to the inhibition of cancer cell growth and apoptosis enhancement [ 159 ]. Furthermore, the overexpression of miR-641 [ 160 ] and mir-644 [ 161 ] led to the inhibition of cervical cancer cell proliferation, migration, and invasion, with a concomitant increase in apoptosis. Ultimately, the ectopic expression of miR-let-7a in SiHa and HeLa cells led to the inhibition of cancer cell proliferation, migration, invasion, and apoptosis induction [ 162 ], while miR-129-5p overexpression in HeLa cells resulted in the inhibition of cell proliferation, increased cell-cycle arrest at the G0/G1 phase, and promoted apoptosis [ 163 ].\nDespite being usually curable following surgery, endometrial cancer still presents as one of the most common female reproductive tract cancer types [ 4 ]. Occasionally aggressive tumours, such as uterine papillary serous carcinomas (UPSC), are observed, with most of them demonstrating aberrant expression of p53 [ 164 ]. Gene therapy strategies toward endometrial cancer involve both viral and non-viral systems, with the former employing primarily adenoviral and retroviral vectors and the latter plasmid DNA/RNA.\nRamondetta et al. performed an adenovirus-mediated expression of  p53  or  p21  in a papillary serous endometrial carcinoma cell line and demonstrated growth inhibition and apoptotic cell death [ 165 ]. Similarly, Ural et al. performed an in vitro suicide gene therapy approach using  HSV-TK  and documented inhibition of endometrial cancer cell growth [ 166 ]. A similar approach, but with the use of the pNF-κB plasmid, along with the gonadotropin-releasing hormone receptor (GnRH-R) agonist triptorelin and the prodrug GCV, resulted in reduced cancer cell growth, both in vitro and in vivo in mice [ 167 ]. Recently, Xia et al. [ 168 ] employed next-generation sequencing (NGS) in four out of the twelve patients enrolled in the clinical trial using the Ad-p53 vector, Genidicine ®  [ 105 ] and observed reduced p53 expression in the tumours of three patients, all carrying mutations in tumour protein  p53  CREB binding protein  (CREBBP),  cyclin-dependent kinase inhibitor 2A  (CDKN2A),  LYN proto-oncogene  (LYN)  and Janus kinase 2  (JAK2)  genes. These data provide strong evidence that NGS can aid in the recruitment of suitable patients for Ad-53 uterine gene therapy.\nEffective gene transfer with a significant inhibition of cancer cells in vitro was also achieved with non-viral approaches. Specifically, Maurice-Duelli et al. demonstrated that the transfer of  PTEN , a frequently mutated gene in endometrial cancer [ 169 ], using PEI-photochemical irradiation could lead to a 44% inhibition in cancer cell growth [ 170 ]. Moreover, the role of certain miRNAs in endometrial cancer has gained a lot of ground, highlighting their potential diagnostic and therapeutic value. In a detailed review, Banno et al. [ 171 ] highlighted the role of several miRNAs differentially expressed in endometrial cancer. Specifically, miR-185, miR-106a, miR-181a, miR-210, miR-423, miR-107, miR-let7c, miR-205, miR-449, and miR-429 were found up-regulated, while miR-let7e, miR-221, miR-30c, miR-152, miR-193, miR-204, miR-99b and miR-193b were significantly down-regulated, suggesting a tumorigenic or tumour-suppressing activity, respectively [ 171 ]. miR-129-2 and miR-152 are involved in the development of endometrial cancer via DNA methylation; miR-125b, mir-30c, miR-200b/c, and miR-429 are related to cisplatin resistance, while miR-125b, miR-30c, miR-194 and miR-34b regulate proliferation, metastasis and invasion of endometrial cancer cells [ 171 ]. Similarly, in a recent systematic review, Donkers et al. [ 172 ] highlighted the importance of miR-205, miR-200 family (miR-200a, miR-200b, and miR-200c), miR-135b, miR-182, miR-183, and miR-223 in endometrial cancer prognosis [ 172 ]. As the above miRNAs are found to be up-regulated in most endometrial cancers, the down-regulation of their expression may have therapeutic outcomes. Regarding miRNAs, such as miR-137, miR-129-3p, and miR-410, whose expression is down-regulated in endometrial cancer, the authors observed little to no consensus [ 172 ], and therefore, their role in both predicting and treating endometrial cancer remains vague.\n\nThey are the most common benign tumours in reproductive-age women [ 173 ] and represent an attractive target for gene therapy, primarily due to their localised nature and slow growth rate [ 5 , 174 ]. The most common gene therapy approaches for fibroid treatment involve suicide gene therapy, antiangiogenic strategy, and mutation compensation and are summarised in  Table 3  and  Figure 5 . The first attempt to employ gene therapy for uterine leiomyoma treatment came from Niu et al. 1998, who used a non-viral approach to deliver the suicide  TK  gene into leiomyoma cells [ 175 ]. Although the authors demonstrated significant cell death despite the low percentage of transfected leiomyoma cells, the above approach could only be applied in vitro. On the contrary, the use of improved Ads [ 174 ] with a leiomyoma-specific expression of critical therapeutic genes, such as  DNER  [ 176 ] and  HSV1-TK  [ 5 ], appear quite promising and present good candidates for human clinical trials. Furthermore, targeted and transduction-efficient Ads, such as Adenovirus-human somatostatin receptor subtype 2-arginine, glycine, and aspartate-thymidine kinase (Ad-SSTR-RGD-TK), given in combination with GCV, show promising results, both in vitro [ 177 ] and in vivo [ 178 ], as they lead to a significant reduction in proliferation and leiomyomas size, the induction of apoptosis, and the inhibition of angiogenic- and extracellular matrix-related genes. Recently, C-terminal Src kinase ( CSK2 ) [ 179 ] and high mobility group AT-hook 2  (HMGA2)  [ 180 ] were highlighted as key players in leiomyomas tumorigenesis. Specifically,  CSK2  was shown to be highly expressed in uterine leiomyosarcomas compared to leiomyomas, and its overexpression was associated with tumour progression and poor prognosis, while  CSK2  silencing using siRNA led to the inhibition of proliferation and cell cycle progression, decreased migration, invasion, and colony formation [ 179 ]. Additionally, the overexpression of  HMGA2  in leiomyomas was shown to be associated with increased vasculature density, demonstrated by the increased expression of angiogenic factors and receptors, such as VEGFA, EGF, bFGF, TGFα, VEGFR1, and VEGFR2, which likely contribute to tumour growth [ 180 ]. Its important role in angiogenesis appears to be through IGF2BP2-mediated pAKT activity [ 180 ].\nIt is a chronic, estrogen-dependent disease characterised by the presence of endometrial tissue outside the uterine cavity, often associated with subfertility [ 5 , 198 ]. In the early 2000s, the above benign gynaecological abnormality made its way to gene therapy applications ( Table 3  and  Figure 5 ). Specifically, Dabrosin et al. using a murine model of endometriosis and employing an adenovirus expressing the murine angiostatin gene managed to completely eradicate endometriotic lesions within two weeks [ 181 ]. A similar antiangiogenic strategy was followed by Sun et al., who used an AAV carrying endostatin (rAAv-endostatin-EGFP) and demonstrated the inhibition of angiogenesis in endometriotic lesions both in vitro and in vivo [ 182 ]. Similarly, but using an endostatin plasmid/PAMAM dendrimer instead, Wang et al. also demonstrated inhibition of endometriosis development following transfection [ 185 ]. Following the above, Othman and co-workers studied the effects of  DNER  gene transfer into endometriosis cells using an Ad-DNER vector and showed that it could induce significant cell death and reduce pro-inflammatory and angiogenic cytokine production by the specific cells [ 183 ]. Both of the above studies used an Ad5 viral vector; however, CRADs, such as Ad-SLP1-luc, Ad-hepanarase-luc [ 184 ], and Ad-sft-1 [ 186 ], achieved higher reporter gene expression and, therefore, may constitute better candidates for the gene therapy of endometriosis. In addition to antiangiogenic strategies, suicide gene therapy and mutation compensation have also been very efficient for endometriosis treatment. More specifically, the use of adenoviral vectors carrying the  TK  gene (Ad-TK) led to significant cell death induction and reduction in endometrial lesions in vivo [ 199 ]. Furthermore, regarding mutation compensation, Goncalves et al. employed Ad-p27, an adenovirus carrying  p27 kip1  coding gene, shown to be down-regulated in women with endometriosis [ 200 ] and demonstrated cell cycle arrest and reduced proliferation [ 188 , 189 ]. Lastly, Paupoo et al., employing CRAd-S-pK7, an adenoviral vector with dual advantages of infection enhancement and promoter specificity, demonstrated high replication rates and cell-killing efficiencies in vitro [ 190 ].\nThe potential contribution of gene therapy for placental abnormalities and functional restoration deficiencies was first introduced by Senut et al. [ 201 ] by injecting genetically modified cells into rat placenta, which demonstrated the increased secretion of gene products in fetal circulation. Furthermore, Xing et al. managed to successfully transfect rat placenta by either systemic administration or intra-placental needle injection of adenoviruses without altering the fetal genome [ 191 ]. However, as intra-placental needle injection may result in fetal-maternal barrier breakdown, Heikilla and co-workers performed a catheter-mediated intravascular gene transfer with adenoviruses, plasmid/liposomes and plasmid/PEI complexes and demonstrated that placental trophoblastic cells could be efficiently transfected with adenoviruses when delivered directly into uterine arteries [ 192 ]. On the contrary, and despite their ability to transfect fetal membranes and the basal plate more efficiently than adenoviruses, plasmid/PEI and plasmid/liposome complexes were not as efficient in transfecting placental trophoblastic cells [ 192 ]. Similarly, the Westwood group successfully used an adenovirus to deliver sense or antisense  IGF-I  or  IGF-II  cDNA into human primary placental fibroblasts (PPF), aiming to study the effects of overexpression or inhibition of these genes on placental cell proliferation, migration, and survival [ 193 ]. The authors reported that placenta cells could be efficiently transfected by adenoviruses, verifying the potential therapeutic outcome of gene therapy in placental disorders. Based on the above, Keswani et al. employed the adenoviral vector Ad-IGF-1 in vivo in mice, demonstrating the restoration of birth weight [ 194 ]. Similarly, Carr et al. showed that using Ad-VEGF led to an improvement in fetal growth and an increased fetal growth rate [ 195 ].\nThe proof of principle that genetic manipulation of the adult uterine endometrium has therapeutic potential against implantation disorders came from Bagot et al. in early 2000 [ 196 ]. The authors first reported that the alteration of maternal Hoxa10 expression, previously shown to be expressed in human endometrium during implantation [ 202 ] by in vivo gene transfection, affects implantation in mice [ 196 ]. Based on this finding, they overexpressed the maternal homeobox A10 ( Hoxa10)  gene using a pcDNA3.1(+)/HOXA10a, a liposome plasmid that constitutively expresses  Hoxa10 , and achieved a significant increase in litter size. All of the mice pups were born following a normal gestation of 17–20 days, were normal in size and exhibited no morphological abnormalities. Similarly, Nakamura et al. [ 197 ], employing the hemagglutinating virus of Japan envelope vector (HVJ), showed that NF-κB expression determines the timing of implantation via control of LIF expression and that manipulation of its expression may alter the implantation outcome. Microarray-based bioinformatics experiments by Horcajadas et al. [ 203 ] led to the identification of a huge number of transcripts regulated during implantation and set the basis for many in vivo gene transfer approaches. Furthermore, a multi-omics approach for the identification of successful implantation biomarkers [ 204 ] can be a useful tool for predicting a successful implantation outcome.\n\nDespite the plethora of published data regarding the pre-clinical evaluation of viral and non-viral gene therapy approaches toward gynaecological cancers, only a small percentage has reached the clinical phase, with most of them not progressing beyond Phase I/II. According to clinicaltrials.gov [ 101 ] and using the terms “ovarian cancer”, “cervix cancer”, and “endometrial cancer” in combination with “gene therapy” as keywords, followed by manual exclusion of trials not involving gene therapy approaches, there are 18 clinical trials for ovarian gene therapy, 11 for cervical gene therapy, and three for endometrial gene therapy, with only a few currently recruiting patients. There are no clinical trials for gene therapy approaches against benign gynaecological disorders.  Table 4  summarises the clinical trials for ovarian, cervical, and endometrial cancer gene therapy.\nRegarding ovarian cancer gene therapy clinical trials, out of the 18 studies, three employed mutation compensation, three suicide gene therapies, and one oncolytic virotherapy strategy, while the remaining 11 used immunopotentiation approaches. In detail, in Phase I  NCT00003450  and  NCT00003588  clinical trials, Ads alone were used to deliver the  p53  gene, while in the Phase II  NCT02435186  study, adenoviral-mediated  p53  delivery was combined with cisplatin and PTX chemotherapy. None of the above clinical trials have been posted or had their results published yet. With regards to suicide gene therapy approaches, clinical trials  NCT00964756 ,  NCT01997190 , and  NCT00005025  employed adenoviral or HSV vectors to deliver the  TK  gene in patients with recurrent ovarian cancer. Published data from Phase I clinical trial  NCT00964756  using the Ad5-SSTR/TK-R-GD vector highlighted its safety and effectiveness, with five out of twelve patients showing stable disease and one out of five demonstrating complete disease resolution [ 205 ]. In the ongoing Phase I/II clinical trial  NCT02068794 , MSCs are used as carrier cells for oncolytic measles virus encoding thyroidal sodium iodide symporter (MeV-NIS) in patients with ovarian cancer.\nRegarding immunopotentiation approaches, in the Phase I clinical trial  NCT00066404 , adenoviral-mediated  IFN-β  transfer induced consistent anti-tumour immune responses; however, the rapid production of anti-adenoviral neutralizing antibodies following second dose administration led to minimized target cell transduction and IFN-β production [ 206 ]. To circumvent the above problem, the authors plan to administer a second dose as early as 3 days after the first one and combine adenoviral administration with chemotherapy or surgery.\nThe rest clinical trials involve administration of engineered T cells expressing either neoantigen-specific TCRs ( NCT03412877 ,  NCT05194735 ,  NCT03970382 ,  NCT04102436 ) or anti-carcinoembryonic antigen (CEA) ( NCT00004178 ), anti-mesothelin ( NCT01583686 ) or folate binding protein ( NCT00019136 ) chimeric TCRs. Specifically, for patients with solid tumours, there are two clinical trials ( NCT02366546  and  NCT02096614 ) assessing the in vivo kinetics and safety of TBI-1301 (NY-ESO-1-specific TCR) and TBI-1201 (MAGE A4-specific TCR), in combination with cyclophosphamide and fludarabine. All the above approaches employing engineered T cells have gained a lot of ground, leading to promising results toward cancer immunotherapy [ 207 ].\nOverall, out of the 18 clinical trials described above, 16 employed viral gene delivery systems, while two, specifically the  NCT04102436  and  NCT00381173  trials, used non-viral approaches, such as the  Sleeping Beauty  transposon/transposase system or plasmid DNA, respectively. Specifically, in the latter study, the administration of ZYC300, a DNA plasmid encoding the tumour-associated carcinogen activator cytochrome P450 1B1  (CYP1B1)  [ 208 ], was used to treat 17 patients with advanced-stage progressive cancer and led to CYP1B1 immunity induction and marked response to next regimen in six of them [ 209 ].\nRegarding the cervical cancer clinical trials, eight out of ten clinical trials employ immunopotentiation approaches, while only two and specifically  NCT03544723  and  NCT03057912 , are based on mutation compensation strategies. Clinical trial  NCT03544723  is a Phase II multi-centre study to evaluate adenoviral-mediated  p53  delivery in combination with immune checkpoint inhibitors, such as anti-PD-1/anti-PD-L1 in patients with carcinomas approved for anti-PD-1 and anti-PD-L1 therapy. Accumulating evidence presented by Liu et al. shows that PD-1/PD-L1 inhibitors may also be a promising therapeutic approach for cervical cancer as well [ 210 ], and therefore the results of the above study (estimated completion date 31/12/2022) will contribute significantly toward efficient cervical cancer treatment. The  NCT03057912  Phase I study assesses the safety and efficacy of the in vivo TALEN- and CRISPR/Cas9-mediated disruption of HPV16- and HPV18-E6/E7 oncoproteins.\nRegarding clinical trials employing immunopotentiation approaches, in the ongoing clinical trial  NCT04180215 , checkpoint inhibitors are administered in combination with vectors expressing HPV16+ antigen (HB-201 and HB-202) to assess their efficacy against patients with metastatic HPV16+ cancers, while clinical trial  NCT00988559  evaluates the safety and efficacy of different routes of administration of pnGVL4a-CRT/E7 (detox) DNA vaccine in patients with HPV16+ cervical intraepithelial neoplasia (CIN)2/3. While the results of the former study will not become available before summer 2025, results of the latter clinical trial indicate induction of CIN2/3 remission following intramuscular or intralesional delivery of pnGVL4a-CRT/E7 (detox) vaccine. The other cervical cancer clinical trials involve the infusion of T cells engineered to express TCRs against HPV16/18 E6/E7 oncoproteins ( NCT02379520  and NCT022280811), DP0401/0402 restricted MAGE-A3 tumour antigen ( NCT02153905 ,  NCT02111850 ), CEA ( NCT00004178 ), or mesothelin ( NCT01583686 ), usually in combination with cyclosphosphamide or fludarabine. The results of the above clinical trials are not published yet; however, the therapeutic potential of the specific approach is quite promising [ 207 ]. Lastly, the anti-tumour effect of  IFN-β  transfer was clearly demonstrated in the  NCT00066404  clinical trial [ 206 ], and it is expected that the revised experimental design will lead to an even more significant therapeutic outcome.\nRegarding endometrial cancer clinical trials, all three ( NCT05194735 , NTC00004178,  NCT00066404 ) employ immunopotentiation approaches discussed above and involve either infusion of CAR T cells or adenoviral-mediated  IFN-β  transfer.\n\nThe aim of this review was to summarise all the major advances regarding gene therapy for gynaecological malignant and benign disorders and highlight their therapeutic potential and their eventual clinical application. It is clear that targeted gene therapy approaches, such as mutation compensation, antiangiogenesis, immunopotentiation, suicide gene therapy, and oncolytic virotherapy, have yielded promising results, employing both viral and non-viral systems. However, most of these strategies either remain in the preclinical phase or show reduced effectiveness as monotherapies and usually require a combination of chemotherapy and radiotherapy to demonstrate a therapeutic outcome. Furthermore, gene therapy in combination with immunotherapy, e.g., targeting CTLA-4 and PD-1, or with the emerging therapeutic angiogenesis inhibitors, may result in increased effectiveness, especially for ovarian cancer in advanced stages, where still no effective therapies exist. Moreover, there is increasing evidence for the therapeutic potential of CAR T-cell technology also, and clinical trials are currently assessing its efficacy, primarily in ovarian cancer, by recruiting patients [ 101 ] ( Table 4 ). The continuous development of safe and effective vectors, the discovery of novel diagnostic markers, such as miRNAs discussed above, or of new players, such as long non-coding RNAs (lncRNAs) [ 211 ] and circular RNAs (circRNAs) [ 212 ] and in combination with more clinical trials beyond Phase I/II are expected to lead to more effective and radical therapeutic outcomes.","source_license":"CC-BY-4.0","license_restricted":false}