Gynecologic Cancer, Cancer Stem Cells, and Possible Targeted Therapies.

CSC surface markers provide molecular targets for several cancers, allowing the use of therapeutic antibodies specific for CSC surface markers. Various CSC surface markers have been identified and published. Interestingly, most of the markers used to identify CSCs are derived from surface markers present on hESCs or adult stem cells ( Kim and Ryu, 2017 ). The development of therapeutic strategies to target CSCs mainly relies on the use of cell surface markers to identify, enrich, and/or isolate CSCs. Many CSC surface markers have been identified, although some surface markers are considered to be controversial and require further investigation. Interestingly, most of the current CSC surface markers are derived from known surface markers of normal embryonic or adult stem cells ( Islam et al., 2015 ; Kim and Ryu, 2017 ). The similarity between cell surface markers suggests that CSCs predominantly originate from normal stem cells via the accumulation of epigenetic and genetic alterations. There are currently 40 published CSC surface markers, which are classified into three different categories, relating to hESCs, adult stem cells, and normal tissue cells ( Kim and Ryu, 2017 ). The first group of CSC surface markers is expressed on hESCs but is weakly or rarely expressed on normal tissue cells. The second group of CSC surface markers is expressed on adult stem cells but is weakly or rarely expressed on normal tissue cells. The third group of CSC surface markers is expressed on hESCs and/or adult stem cells, and is also considerably expressed on several normal tissue cells. CD133 is the most frequently studied CSC surface marker in many cancers, and specific antibodies/immunotoxins against CD133 have been successfully developed for the selective eradication of CSCs ( Bach et al., 2013 ; Schmohl and Vallera, 2016 ). CD133 is also one of the most frequently studied surface markers in solid cancers. It appears that the level of CD133 protein expression does not alter upon cell differentiation; however, tertiary conformational changes in differentiated colon cancer cells block the binding of an anti-CD133 antibody, suggesting that the expression of the CD133 epitope is restricted to undifferentiated stem cells ( Grosse-Gehling et al., 2013 ). CD117 is involved in signal transduction for survival and self-renewal in various cells ( McLaughlin et al., 2007 ). Human epithelial OC CD44+CD117+ cells were shown to possess CSC properties, along with increased chemoresistance ( Chen et al., 2013 ). LGR5 is a CSC marker in mouse intestinal cancer ( Ahmed et al., 2010 ) and has also been suggested to be a CSC marker for human colon tissue and colorectal cancer ( Kemper et al., 2012 ; Hirsch et al., 2014 ).

Angiogenesis is a process that is crucial in cancer progression. During the tumorigenesis process, the balance between proangiogenic factors and angiogenic inhibitors becomes disturbed. One of the most important regulators of angiogenesis is vascular endothelial growth factor A (VEGF-A), which is involved in the progression of the epithelial ovarian cancer. Accordingly, some drugs have been developed including bevacizumab (a monoclonal anti-VEGF antibody) to inhibit angiogenesis during cancer progression ( Burger et al., 2011 ). In another therapeutic approach, neoadjuvant chemotherapy after interval debulking surgery has been used to manage gynecologic cancer, especially advanced ovarian malignancies ( Chéreau et al., 2013 ). Some studies have reported that the expression and activity of matrix metalloproteinases (MMPs) are important in several human cancers. These proteins contribute to invasion, metastasis, and the advanced stage of the tumor ( Hua et al., 2011 ). MMP-9 and MMP-2 were increased in invasive endometrial cancer, and the high amounts of MMP-9 and MMP-2 colocalized with ETV5/ERM and RUNX1/AML1 (factors associated with neoplastic progression) ( Planagumà et al., 2011 ). Another event that may occur during cancer development involves changes in cell–cell adhesion molecules, resulting in increased susceptibility of cells to exfoliation. One of the main molecules contributing to the adherence of adjacent cells is E-cadherin, a glycoprotein located at the cellular adherent junctions. In ovarian cancer, the expression level of E-cadherin in floating cancer cells in ascites and in metastatic deposits was lower than that in primary ovarian tumors ( Sawada et al., 2008 ). The “seed-and-soil” hypothesis to explain the metastasis process was first attributed to Stephan Paget (1889), who proposed that circulating cancer cells (seeds) were only able to metastasize to organs where the microenvironment was particularly suitable for their growth and development ( Zhang et al., 2010 ). C-X-C chemokine receptor type 4 (CXCR4) (also known as CD184) is a diagnostic marker in several cancers, including breast cancer. The CXCR4 gene encodes a receptor protein located in the cell membrane, through which signaling pathways are activated, followed by regulation of cell proliferation. Current advances in cancer biology have highlighted the crucial role of CXCR4 receptor and its respective ligand CXCL12, in the metastasis of different kinds of cancer ( Mukherjee and Zhao, 2013 ). In different human cancers, the expression of the estrogen-responsive gene RCAS1 has been correlated to clinical outcomes in ovarian cancer and to the survival rate in patients suffering from uterine and cervical adenocarcinoma, as well as esophageal, pancreatic, lung, gallbladder, and pancreatic cancers. Researchers have also shown a relationship between the expression of RCAS1 and the invasion and metastasis of cervical, stomach, skin, breast, and thyroid cancers. RCAS1 contributes importantly to the aggressive behavior of several kinds of cancer ( Sonoda et al., 2005 ). Moreover, the transcription factor Snail potentially contributes to the epithelial–mesenchymal transition (EMT) process and to cancer growth, metastasis, and invasion. It has been shown that MMP expression in mammary epithelial cells can stimulate Snail expression and subsequently the EMT in cancer cells ( Przybylo and Radisky, 2007 ). Moreover, RCAS1 promotes angiogenesis and accelerates tumor growth in immune-deficient nude mice. Taken together, the data suggest that RCAS1 affects the tumor–stroma interaction to enhance angiogenesis and is crucial for tumor growth in vivo . The expression of VEGF can be affected by RCAS1 , thus stimulating angiogenesis, endothelial cell motility, and vascular permeability ( Sonoda et al., 2007 ). Moreover, the transforming growth factor-β (TGF-β) contributes to cancer progression, angiogenesis, escape from immunosurveillance, and myofibroblast recruitment. PI3K or phosphoinositide 3-kinase is a family of the enzymes that contributes to cellular functions like proliferation, motility, differentiation, intracellular trafficking, and cell survival. Previous studies have shown that the expression of PI3K is increased in ∼40% of ovarian cancers, while the expression and activity of its downstream effectors, AKT2 and AKT1, are higher in diverse cancers, such as ovarian tumors. In several kinds of tumor cells, AKT1 increases the stability of hypoxia-inducible factor, HIF-1α ( Liby et al., 2012 ). During the angiogenesis process, researchers have observed the degradation of the basement membrane matrix by proteinases, among which the membrane-type matrix metalloproteinase (MT-MMP) is the most important example. As the angiogenesis process develops, the extracellular matrix (ECMs) proteins affect signaling cascades that contribute to proliferation, invasion, migration, and survival ( Davis and Senger, 2005 ). In Table 1 , the changes in genes and proteins in different kinds of gynecologic cancers are summarized. In the following sections, the main molecular mechanisms involved in female genital malignancies are discussed to provide more information to choose the best therapeutic approaches. Proteins and genes that have roles in gynecologic malignancies.

Many efforts have been made to design therapeutic approaches that specifically target CSC populations. This is because CSCs have been recently predicted to be a crucial population to eliminate. However, recent insights have complicated the initially elegant model, by showing the dominant role for the tumor microenvironment in determining CSC properties. This is particularly important since the dedifferentiation of non-tumorigenic cancer cells to produce CSCs has been shown to occur, and therefore, the CSC population in a neoplasm can vary over time. Moreover, evidence suggests that not all tumors are driven by rare CSCs, but might instead contain a larger population of tumorigenic cells. Even though these results suggest that specific targeting of the CSC population might not necessarily be a useful therapeutic strategy, research into the hierarchical cellular organization of tumors has provided many important new insights into the biology of tumors. Due to the deficiencies of conventional routine therapy of gynecologic cancer, especially recurrent and advanced stages, the development of specialized and more targeted therapeutic approaches could be very promising. Furthermore, accumulating data concerning genomic and proteomic profiling over the last several years have paved the way for understanding the molecular foundation of human cancer and the role of several genes, which have been altered, activated, or inactivated in tumors. Therefore, the present review has summarized some advances in the targeted treatment for each molecular tumor profile. Importantly, CSCs are one of the current limitations in treatment of different cancers because of their resistance to chemotherapy drugs caused by different mechanisms. Therefore, more studies are needed to target these cells and to understand their functional mechanisms to prevent the recurrence of cancer.

As mentioned previously, multiple preclinical and clinical reports that demonstrate not only the significance of CSCs in different types of cancer, metastasis, and recurrence but also suggest that targeted therapies against these cells may be effective ( Desai et al., 2019 ). As shown Figure 2 , a small number of CSCs within the total tumor burden is capable of reproducing the whole tumor mass, so that targeting these cells in addition to conventional therapy could help prevent tumor recurrence. Effectiveness of targeted CSC therapy in combination with conventional chemotherapy. Conventional chemotherapy can decrease tumor size by acting on non-CSC tumor cells; however, tumor recurrence associated with CSCs may lead to treatment failure. Targeting the CSCs could reduce numbers of CSCs followed by better response of tumors to therapy. (The figure was reprinted from Kyo’s (2013) study.) CSCs are extremely important in OC, and their targeted elimination may be an efficient approach to reduce resistance to chemotherapy and cancer recurrence ( Walters Haygood et al., 2014 ). Targeting signaling pathways in OC could be used to attack CSCs. Because these signaling pathways are responsible for the increased survival of CSCs, disrupting the pathways could help eliminate the CSCs. In fact, the NOTCH signaling pathway contributes importantly to the maintenance and survival of CSCs ( Barnawi et al., 2016 ). Another targeted therapy for ovarian CSCs could rely on the elimination of cell surface markers like CD44, CD24, CD133, and CD117. It was reported that CD133 + OVCAR5-luc cells could be eliminated, which significantly reduced tumor progression ( Skubitz et al., 2013 ). Moreover, there are other treatment strategies available for the targeted elimination of ovarian CSCs relying on signaling pathways, surface markers, interfering with niches, miRNAs, or differentiation therapy ( Keyvani et al., 2019 ). Several CSC-specific markers and signaling pathways could be treatment targets for CSCs, although they have not been extensively studied in CC. Moreover, several therapeutic methods have been proposed and evaluated to attack CSCs ( Sudhalkar et al., 2019 ). One new therapeutic approach involves CSC-targeted nanoparticles (NPs), in which the NPs are designed to affect stem cell-associated functions via targeting specific signaling pathways such as Notch, reactive oxygen species (ROS) signaling, CSC-specific markers, or Wnt/β-catenin. Also, these NPs may interfere with the maintenance of cell stemness ( Wang et al., 2012 ; Yang et al., 2014 ; Dou et al., 2015 ; Hong et al., 2015 ; Kim et al., 2015 ; Rao et al., 2015 ; Rotherham and El Haj, 2015 ). Similarly, the targeting of endometrial CSCs could be used to improve the chance of a complete cure and prevent its recurrence. There are various strategies including targeting the surface markers of the cancer stem cells, including CD133+ and CD117+. Different molecular techniques can be used for targeting of these cells including molecular-targeted agents such as miR-199b-5p, the γ-secretase inhibitor DAPT, and inhibitors of the Notch pathway ( Garzia et al., 2009 ; Hovinga et al., 2010 ). Some studies have reported the proapoptotic and antiproliferative effects of HDAC inhibitors on EC cells ( Takai et al., 2004 ; Jiang et al., 2007 ; Ahn et al., 2008 ). Moreover, salinomycin has been shown to inhibit fibronectin expression and reduce the proliferation, migration, and invasion of endometrial CSCs (RK12V-SP and Hec1) ( Lu et al., 2011 ).

At least 3–6 separate genetic modifications are needed to transform a normal cell into a cancerous cell ( Croce, 2008 ). Most cancer cells are genetically unstable, and this instability results in the accumulation of numerous secondary molecular changes in the cell, which play important roles in the development of the malignant properties, like invasion, immortality, drug resistance, and metastasis ( Hanahan and Weinberg, 2011 ). Genomic instability is associated with a higher frequency of mutations in the genome. These mutations can be very different and can include alterations in the sequence of nucleic acids, chromosomal rearrangements, or aneuploidy. Genomic instability has particular importance in multicellular organisms and can be the cause of neurological diseases like neurotoxic myotonic dystrophy or amyotrophic lateral sclerosis ( Schmitt et al., 2012 ). This instability may be due to the higher frequency of external DNA damage followed by mutations caused by errors in the repair process or by incorrect translation. Other sources of genomic instability include mutational and/or epigenetic reductions in the expression of DNA repair genes. Endogenous DNA damage is common; indeed it occurs >60,000 times/day in the human cellular genome ( Møller, 2005 ). Certain modifications in genes stimulating cell growth (oncogenes) may also lead to malignancy in normal cells ( Doroshow and Kummar, 2014 ). Oncogenes are activated by different mechanisms. Upregulation of oncogenes can increase the expression of the proteins, while one-point mutations can lead to oncogene activation. In addition, oncogenes can be shifted from one chromosome to another and can be affected by the promoter region of the new site, resulting in the enhancement of oncogene expression ( Negrini et al., 2010 ). Tumor-related oncogenes can cause unscheduled cellular proliferation and also chromosomal and genomic instability. There are a number of treatments that can not only block the activity of oncogenes but also specifically target tumor cells. Nevertheless, some studies have shown the involvement of oncogenes in the rapid growth of certain tumor cells, but not all types of tumors. Furthermore, oncogenes and their target cells interact with each other through the induction of proliferation and developmental reprogramming of the epigenome, which contributes essentially to the expansion of tumors. Evidence suggests a possible initiation of tumorigenesis through stem cell reprogramming, which could be a new role of oncogenes in the tumorigenesis process ( Vicente-Dueñas et al., 2013 ). Another issue during tumorigenesis is the upregulation of cell rapid growth modulated through cell cycle-associated molecules like RB, TP53, cyclins, CDKs, and E2F ( Harris and Levine, 2005 ). Cell cycle dysregulation is one of the most important events that could occur during tumorigenesis, in which cells become highly resistant to senescence and cell death. In contrast, malignant cells undergo a cellular senescence process during normal cell cycle progression ( Collado and Serrano, 2010 ). Cell death has different types, defined by various morphological criteria, including apoptotic, necrotic, autophagic, or mitosis-associated cell death ( Su et al., 2015 ). Further information on the molecular basis of gynecologic cancer and therapeutic options, based on CSCs, will be addressed in the next section.

Cancer is a group of diseases associated with the abnormal growth of malignant cells and their expansion to other areas of the human body ( Grever et al., 1992 ). Benign types of tumors are different from malignant tumor types and do not show the same metastatic activity. Cancer is caused by genetic and epigenetic modifications that enable uncontrolled cell growth, migration, and disturbances in cell death pathways. There are a number of molecular changes that drive the transformation of normal cells into malignant cells, but the spectrum and diversity of these changes vary widely among different cancer types ( Spandidos et al., 2000 ). Gynecologic malignancies, including ovarian, cervical, and endometrial cancer, seriously affect the health of women worldwide, contributing considerably to the global cancer burden. Epithelial ovarian cancer comprises ∼90% of malignant ovarian neoplasms, which is a leading cause of death in women. The 5-year overall survival (OS) rate of OC is ∼47% for all stages, and >70% of patients are diagnosed at an advanced stage with an even lower 5-year OS rate ( Wang et al., 2020 ). Currently, some potential therapeutic targets include tumor-intrinsic signaling pathways, angiogenesis, homologous recombination deficiency (HDR), hormone receptors, and immunologic factors. The corresponding targeted therapeutic agents include signaling pathway inhibitors, antiangiogenic agents, poly (ADP-ribose) polymerase (PARP) inhibitors, selective estrogen receptor downregulators, and immune checkpoint inhibitors ( Wang et al., 2020 ). In 1877, Virchow’s student Cohnheim discovered a new cell population in tumors and pointed out that it possessed an embryonic character ( Spandidos et al., 2000 ). Today, those cells are called cancer stem cells (CSCs) or tumor-initiating cells (TICs) and are seen as drivers of tumor establishment and growth, often correlated to aggressive, heterogeneous, and therapy-resistant tumors. CSCs are a subpopulation of tumor cells that can drive tumor initiation and can cause recurrence after treatment. At the time point of tumor initiation, CSCs can originate from either differentiated cells or adult tissue-resident stem cells. Due to their importance, several biomarkers to characterize CSCs have been identified and correlated with diagnosis, prognosis, and response to therapy in patients. However, CSCs have been shown to display a high degree of plasticity, which changes their phenotype and functional properties. Such changes are induced by chemotherapy and radiotherapy, as well as the presence of senescent tumor cells, which cause alterations in the tumor microenvironment ( Walcher et al., 2020 ). Stem cell studies have provided scientists with useful information about understanding the contribution of the CSCs to cancer progression, resulting in the discovery of novel methods and therapeutic targets of CSCs ( Meacham and Morrison, 2013 ). CSCs possess the capacity for self-renewal and can generate heterogeneous lineages of cancer cells within tumors. A substantial body of evidence supports a model in which CSCs play a major role in the initiation, progression, and clinical outcome of cancer. The initiation of cancer by CSCs is attributed to their stemness property, allowing them to accumulate underlying carcinogenic mutations including those related to inflammation and oxidative stress. CSCs can further promote cancer growth and progression by mutual interaction with the microenvironment, which allows them to favor their own survival, expansion, resistance to therapy, promotion of angiogenesis, and metastatic capability. Therefore, CSCs are potential therapeutic targets for the development of therapies that could control cancer and achieve improved clinical responses in patients ( Ayob and Ramasamy, 2018 ). The CSC model proposes that tumor initiation, growth, and progression are fueled and sustained by undifferentiated cancer cells endowed with self-renewal on the one hand and more differentiated cells on the other hand. Gynecologic malignancies, based on their biological behavior and clinical course, represent a typical example of CSC-driven cancer. For example, several markers such as CD133, ALDH1/2, LY6A, LGR5, EpCAM, CD133, CD44, CD34, CD24, CD117, MyD88, and CDH1 have been used for the isolation of CSCs from ovarian cancer cell lines. The most common signaling pathways activated by endometrial CSCs are Wnt/β catenin, Notch1, and Hedgehog. Targeting the Notch3 pathway (a transmembrane protein) is a novel method for possible eradication of ovarian CSCs. Gamma-secretase inhibitors (GSI) active against Notch receptors have been used in preclinical and clinical studies ( Zhan et al., 2013 ; Moghbeli et al., 2014 ; Keyvani et al., 2019 ). The use of CSCs has been evaluated as vaccines, for example, in colorectal cancer ( NCT02176746 ), hepatocellular cancer ( NCT02089919 ), and pancreatic cancer ( NCT02074046 ). Traditional cancer therapies, including routine chemotherapy and radiotherapy, have major limitations, especially in eradication of CSCs, leading to frequent recurrence of the cancer mass. Therefore, targeting CSCs could be highly effective in preventing cancer recurrence ( Dragu et al., 2015 ). In this regard, mastermind-like 1 (MAML1), a molecule with few side effects, may be used for targeting CD44 + CSCs via repressing the canonical NOTCH pathway in esophageal squamous cell carcinoma (ESCC) patients ( Moghbeli et al., 2019 ). CSCs are slow cycling and are capable of both self-renewal and differentiation. In the context of time-dependent tumor growth, CSCs actively participate in tumor mass expansion and morphogenesis. Many articles have shown that CSCs contain a spectrum of heterogeneous cell populations. These so-called “stem cell markers” can also change with time. Moreover, CSCs show a great variation in percentage composition across different tumor types. Thus, to claim that CSCs persist in a non-dividing state may not be accurate. Perhaps, CSCs can migrate to the bone marrow or other organs where they may remain in a dormant state for years. The molecular mechanisms underlying gynecologic cancers, the specific molecular changes in these cancers, and the contribution of CSCs to the expansion and survival of these cancers are reviewed, in order to highlight the importance of targeting CSCs as a new therapeutic approach in the field of personalized medicine.