Ovarian cancer (OC) is considered the most lethal gynecological neoplasm due to the high number of deaths in proportion to its incidence. The latest report from the International Agency for Research shows that in 2020 it was the cause of 313,959 new cases and 207,252 deaths in the world (2). The lack of laboratory tests and the absence of pathognomonic symptoms that evidence the early stages of OC, cause this neoplasm to be diagnosed in late stages (III and IV), when cancer has already metastasized, compromising other organs and generating a poor prognosis. The low 5-year survival rate of patients with late-stage epithelial ovarian cancer (30%) is a consequence of the rapid progression of this disease (100). OC is promoted by factors of the TME, among them CAFs stand out. CAFs are the predominant cell type in the TME of different neoplasms; in OC their presence is associated with increased migration and invasion of cancer cells (101–103), extracellular matrix remodeling (104), spheroid formation (105), tumor growth promotion (106), metastasis (107, 108), angiogenesis (109), chemoresistance (110) and poor prognosis for OC patients (111, 112), cellular events that promote neoplastic development and make them a good therapeutic candidate.

In OC, as in other neoplasms, it has been shown that TGF-β plays an important role. This growth factor is secreted by cancer cells, inducing the differentiation of normal fibroblasts into CAFs. The sphingosine kinase 1 (SphK1) enzyme is responsible for mediating TGF-β signaling through the transactivation of sphingosine 1-phosphate receptor 2 (S1PR2) and S1PR3, leading to p38 MAPK phosphorylation and the consequent acquisition of the CAF phenotype. In vivo assays demonstrated the participation of SphK1 in tumorigenesis, promoting tumor growth and metastasis (113). CAFs can also secrete TGF-β, which activates paracrine signaling, promoting the progression of OC due to the expression of various genes, including MMP11 and MMP13, promoting the metastasis of ovarian cancer cells to adjacent or distant sites. Overexpression of MMP13, along with CGA, EPHA3, PSMD9, PITX2, and PHLPP1, has been associated with poor response to platinum-based chemotherapy in patients with high-grade serous ovarian cancer (HGSOC), the most aggressive form of ovarian cancer (114). Li et al. conducted a study with CAFs isolated from stage IIIC OC patients in which they determined that CAFs, through the secretion of exosomes loaded with TGF-β, can promote various pro-tumor events such as migration, invasion, and EMT of OC cells through the activation of SMAD signaling pathway. Additionally, in a model of xenotransplanted mice, they observed that the co-inoculation of ovarian cancer cells and CAFs favored the generation of tumors compared to animals inoculated only with ovarian cancer cells (30), these results suggest the pro-tumor role of CAFs and TGF-β in the TME of OC.

CAFs promote tumor development by providing the cells of the TME with the necessary signals to proliferate and enhance angiogenesis, through the overexpression of NF-kB, IL-6, cyclooxygenase-2 (Cox-2), and CXCL1, molecules known to have pro-tumorigenic activity (115). It has also been observed that CAFs can overexpress nicotinamide N-methyltransferase methyltransferase (NNMT), an epigenetic regulatory molecule that acts through hypomethylation of DNA, RNA, or histones via attenuation of the S-adenosyl methionine/S-adenosyl homocysteine (SAM/SAH) ratio. This enzyme is essential for the expression of CAF markers, the secretion of pro-tumorigenic cytokines, and oncogenic ECM. CAFs overexpressing NNMT promoted OC migration, proliferation, EMT, and metastasis (116).

CAFs not only maintain cellular communication with cancer cells, but it has also been observed that in the TME of OC, communication between CAFs and cancer-associated macrophages (CAMs) leads to cancer cell invasion and metastasis. CAFs, through secretion of IL-33, lead to the expression of M2 macrophage marker genes in human blood‐derived monocytes; in turn, CAF‐induced CAMs increase the invasion and migration of OC cells and upregulate the EMT marker genes (117). To promote tumor progression, CAFs also maintain communication with endothelial cells, upregulating the lipoma-preferred partner (LPP) gene in microvascular endothelial cells (MECs); this event increases the formation of focal adhesions and stress fibers and, consequently, endothelial cell motility, intratumoral microvessel leakiness and chemoresistance to paclitaxel. Mechanistically, MFAP5 derived from CAFs promoted the expression of LPP in microvascular endothelial cells by activating the focal adhesion kinase/extracellular signal-regulated kinase/cAMP responsive element binding protein (FAK/ERK/CREB) signaling network, generating resistance to paclitaxel by increasing cell migration and their focal adhesions, which weakens blood vessels causing paclitaxel to leak before being released to OC cells (118). MFAP5, overexpressed and secreted by CAFs into the EMC, binds to the αvβ3 integrin of OC cells. This ligand-receptor union activates the calcium-dependent FAK/ERK/CREB signaling pathway, leading to overexpression of troponin C type 1 (TNNC1), increasing cell mobility by promoting the formation and rearrangement of the F-actin cytoskeleton. TNNC1 has been proposed as a biomarker for poor prognosis for HGSOC. In vivo assays, MFAP5 promoted tumor progression by increasing angiogenesis, tumor growth, invasion, and metastasis (84) and MFAP5 blockade using monoclonal antibodies was able to inhibit fibrosis and enhance chemosensitivity in mouse models, generating tumor suppression (119). These results propose MFAP5 as a new and promising therapeutic target in patients with OC.

FAP, a classic marker of CAFs, has been associated with recurrence in patients with epithelial OC after treatment with neoadjuvant chemotherapy; strongly FAP-positive tumor parenchyma and stroma were seen in tumors from patients with high recurrence rates (OR: 15.95) while FAP-negative tumors were seen in patients with lower recurrence rates (OR: 0.086) (120). FAP could be used clinically as a negative prognostic marker for patients with OC. At the same time, it could be the target of inhibitory molecules that help counteract its pro-tumor action in the TME.

CAFs manifest their pro-tumorigenic behavior across a broad molecular spectrum; myristoylated alanine-rich C-kinase substrate (MARCKS) is overexpressed in CAFs and has been strongly associated with expression of the classic CAF marker α-SMA. MARCKS participates in the activation, proliferation, chemotherapeutic resistance, and migration of CAFs, its silencing in CAFs decreases the proliferation and migration of OC cells and tumor growth, together MARCKS derived from CAFs facilitates OC metastasis (121), for which it is suggested that MARCKS could be an attractive target for the therapy of patients with OC. Similarly, periostin (POSTN) is enriched in the OC stroma, particularly in CAFs from HGSC; its overexpression was correlated with reduced overall survival. POSTN increased OC cell migration and invasion by functioning as an αvβ3 integrin ligand, activating the PI3K/Akt pathway and promoting the EMT (122); stromal-derived POSTN may be a potential therapeutic target given that it participates in the remodeling of the pre-metastatic niche (PMN) of OC. CAF-derived CXCL12 also induces EMT via the CXCR4/Wnt/β-catenin pathway; CXCL12 expression was associated with cisplatin chemoresistance in OC patients (123). These findings, together with other evidence, propose CAFs as a modulator cell type with a decisive role in generating chemoresistance, angiogenesis, remodeling, and immunomodulating the TME and the PMN (124–126).

Cancerous development has a high energy demand, glucose being the main molecule to obtain ATP and the high consumption of glucose allows cancer cells to maintain their high rate of uncontrolled growth. CAFs favor glucose uptake by OC cells modulating the activation of key enzymes in cellular glycogen metabolism, which increases the energy available to enhance the migration and invasion of OC cells. In human samples, it was observed that the greater the progression of the OC, the lower the glycogen level in the tissue and the greater the number of CAFs. The authors demonstrated that glycogen mobilization in cancer cells depended on p38α MAPK activation in CAFs by TGFβ released by the cancer cells, leading to increased secretion of the cytokines CXCL10, IL-6, and IL-8 (127, 128).

The difficulty of diagnosing ovarian cancer in its early stages is mainly due to the anatomical site where the ovaries are located and the lack of early biomarkers; in search of new candidates for biomarkers, Lawrenson et al., have non-invasively detected natriuretic peptide B (NPPB), a hormone initially unrelated to ovarian neoplasia and that is secreted from cardiac ventricular myocytes in response to myocardial stretch and stress, in the stroma of 60% primary OC tissues and the blood of 50% of women with OC (129). Although NPPB was expressed by only 28% of early-stage OCs, it could be part of the design of a panel of early biomarkers that together increase their sensitivity and specificity.

There are several reports in which it is shown that CAFs non-coding RNAs (ncRNAs) promote the development of various cancers (130). A predictive functional analysis identified 39 lncRNAs differentially expressed in CAFs compared to fibroblasts, 17 lncRNAs were up-regulated in CAFs and 22 lncRNAs were downregulated. The in silico analysis allowed to associate this signature of lncRNAs with multiple pathways in OC metastasis (131), the findings of this study suggest that lncRNAs modulate the CAFs phenotype and that the upregulated or deregulated expression of these ncRNAs favors OC metastasis and progression. An example of deregulated miRNA in CAFs is miR-124, a highly conserved tumor suppressor ncRNA, and highly expressed in fibroblasts. Downregulation of miR-124 is associated with tumorigenesis, tumor progression, and poor prognosis in OC patients. Its decrease or inhibition promotes cell migration, which leads to the acquisition of an aggressive cell phenotype. The cellular effects of miR-124 are exerted by inhibiting SphK1, which catalyzes the phosphorylation of sphingosine to form sphingosine 1-phosphate (SPP), a key sphingolipid signaling molecule involved in cell growth, survival, differentiation and motility (132). Additionally, it has been observed that exosomes loaded with miR-98-5p secreted by CAFs increase cell proliferation, inhibit apoptosis, and promote cisplatin resistance in ovarian cancer cells by downregulating cyclin-dependent kinase inhibitor 1A (CDKN1A) (133).

Despite the extensive evidence on the pro-tumor role of CAFs in the TME, favoring tumor growth and metastasis (134, 135), there are some reports that suggest an anti-tumor role for these cells of heterogeneous origin. Colvin et al. demonstrated that CAFs are capable of secreting ncRNAs with anti-tumor properties, such as MIR155HG. This lncRNA could be considered as a biomarker of good prognosis for patients with OC, since its overexpression has been associated with more remarkable patient survival, T cell activation, antigen processing and presentation, and with the enriched infiltrate of immune cell anti-tumor within the tumor, highlighting plasma cells, various subpopulations of T cells, M1 macrophages, and eosinophils (136). This finding opens the possibility to the existence of subpopulations of CAFs with a clear anti-tumor role; modulating, targeting, or enriching these cell subpopulations in the TME could help fight early-stage cancer development.

Fibroblasts are present in the endometrial stroma, which surrounds the endometrial glandular epithelium. These fibroblasts can be transformed into CAFs by endometrial cancer cells, and once this transformation occurs, CAFs gain some characteristics that support the development of cancer (137). The accumulating evidence shows that CAFs can communicate with microenvironment cells through extracellular vesicles (EV)/exosomes, which transport several molecules, including miRNAs (138). Some miRNAs have a regulatory activity, either pro- or anti-tumor; being the latter the case of miR-320a. In endometrial cancer (EC), miR-320a inhibits the proliferation of EC cells, downregulates the expression of hypoxia-inducible factor 1 alpha (HIF-1α) in EC cells, and inhibits HIF1α/VEGFA axis (139). This axis, in particular, is fundamental in EC because its inhibition is suggested to have an improved radiosensitivity in EC cells (140). Even though CAFs can directly transfer miR-320a into EC cells, they normally express this tumoral suppressor miRNA to a lower extent (139), encouraging debate on the pro- or anti-tumor role of CAFs. miR-148b has also been reported to have anti-tumor effects. Li et al. concluded that loss of exosomal miR-148b promotes metastasis in vitro and in vivo, and it induces EMT and invasion of endometrial cancer cells via the relief of the suppression of DNA methyltransferase 1 (DNMT1) (141). However, other miRNAs have been reported to be associated with important features of cancer, like miR-22, which has been associated with resistance to tamoxifen treatment in patients with breast cancer (142). Taken altogether, miRNAs in EV derived from CAFs could be evidencing the existence of diverse populations of CAFs, both anti- and pro-tumor depending on the type and amount of miRNAs they contain, that inhibit or promote events which favor malignancy development, making them a focus for targeted treatment in EC.

A new approach for therapeutics is directed to the interaction between lncRNAs and miRNAs, also called Competing Endogenous RNA (ceRNA) (143). Several reports suggest that this regulation may be vital for CAFs to support growth and metastasis in endometrial cancer. Long non-coding RNAs have also been involved in EC development, such as the case of nuclear enriched abundant transcript 1 (NEAT1), which was transferred to the endometrial cell lines HEC-1A and RL95-2 through the exosomes derived from CAFs and promoted tumorigenicity in vivo, increasing the levels of YKL-40, an EC marker. The opposite effect was observed by miR-26a/b-5p, which decreased the YKL-40 and STAT3 expression. NEAT1-overexpressing CAFs downregulated the expression of miR-26a/b-5p, and the opposite effect was observed by CAFs that did not express NEAT1 (25); this might represent an example of pro-tumor ceRNA, it is necessary to design and carry out clinical studies where its anti-tumor uses are evidenced by inhibiting miRNAs that promote cancer progression. Studies suggest that PTTG is a key modulating factor in carcinogenesis because of its involvement in processes like cell cycle progression, angiogenesis, malignant transformation, and metastasis (144, 145). Wang et al. reported that CAFs could induce the expression of PTTG in EC and therefore contribute to the carcinogenesis progress. The pituitary tumor transforming gene (PTTG) expression was evaluated by RT-PCR and Western blot showing a significant increase in the mRNA and protein levels in the EC cells cocultured with CAFs compared to those with normal fibroblasts. They also observed that CAFs increased EC cell invasion and migration and induced EMT in EC cells by regulating PTTG in an in vitro model (146). CAFs derived from human endometrium induced a higher proliferation of endometrial cells than normal endometrial fibroblasts. AKT and ERK were highly phosphorylated in ECC-1 cells after being treated with medium conditioned of CAFs, which might indicate that PI3K and MAPK are responsible for the observed proliferation effect (147).

Similar to other cells, CAFs can respond to hormones. Estradiol (E2) and progesterone are essential hormones in regulating the menstrual cycle and mitogenic responses. Unopposed E2 can lead to endometrial hyperplasia due to its proliferative effects, while progestins are protective, leading to cell differentiation of endometrial glands, in contrast to E2; in fact, synthetic progestins are indicated in the treatment of complex atypical hyperplasia and low-grade type I EC, suppressing the actions of E2 (148–151). TME of EC has been documented to express low levels of progesterone receptor and estrogen receptor alpha (ER-α) (152), which may cause the lack of responsiveness of the CAFs to these hormones. Besides, this may lead to altering the paracrine regulation in EC. Progesterone induces the release of PEDF in CAFs and endometrial stromal fibroblasts, the expression of this potential growth inhibitor is associated with several human cancers; whereas high concentrations relate to a good prognosis, low ones are known to be pro-tumorigenic (153).

Another reported effect of CAFs is the suppression of natural killer (NK) cells´ cytotoxic activity. Inoue et al. reported a decrease in the killing activity of NK cells, which was cell-cell contact-dependent and associated with a diminished cell-surface Poliovirus receptor (PVR) expression, an essential ligand for DNAX accessory molecule-1 (DNAM-1), an activating receptor present on NK cells (137). Finally, CAFs may support EC development by the secretion of proinflammatory cytokines such as IL-8, IL-6, monocyte chemoattractant protein-1, and VEGF, and the tumoral growth stimulation through the stromal cell-derived factor–1a/CXC chemokine receptor 4 axis (147, 154, 155). More research must be conducted to identify the CAF secretome in the EC context.

CAFs have fundamental participation in the progression of cervical cancer (CC); cancer cells and CAFs have paracrine communication through various soluble factors such as proteins, IncRNAs, and miRNAs, among other molecules. Wnt2B is a protein secreted by cancer cells in exosomes, which participates in the differentiation of fibroblasts towards CAFs, activating the Wnt/β-catenin signaling pathway, inducing transcription through LEF/TCF and, consequently, promoting tumor growth in nude mice (156). In CC, one of the generating factors of CAFs is IncRNAs; homeodomain-interacting protein kinase 1 antisense RNA (HIPK1-AS) overexpressed in HeLa conditioned medium induced the differentiation of normal human cervical fibroblasts towards CAFs, increasing the expression of FAP, IL-6 and α-SMA; HIPK1-AS modulation is reversible because its deletion down-regulates CAF activation; the mechanisms of HIPK1-AS are not yet fully elucidated (157). In addition to being a CAFs-generating molecule, TGF-β is secreted by itself, promoting migration and invasion of CC cells, essential events for tumor growth and metastasis (28, 158, 159). These findings show the important communication between cancer cells, fibroblast, and CAFs to facilitate metastasis in CC.

In the TME of the CC, the two-way communication between CC cells and CAFs is constant and favors the development of the disease. PDGF-BB secreted by cancer cells induces the expression of heparin-binding epidermal growth factor-like growth factor (HB-EGF) by CAFs, through activation of PDGFRβ, which in turn activates epidermal growth factor receptor (EGFR) in CC cells promoting cell proliferation and tumor growth, in in vitro and in vivo models, respectively. In patient´s tissues, it was observed that HB-EGF expression is increased during cervical carcinogenesis (160), suggesting that it participates in carcinogenesis, maintenance, and progression of the CC.

microRNAs contained in extracellular vesicles derived from CAFs facilitate tumor growth in in vivo models, such as the case of microRNA-10a-5, which is capable of inducing angiogenesis via activating the Hh pathway by inhibiting TBX5 and promoting the expression of VEGF (18). In athymic nude mice, it was observed that the co-inoculation of CAFs and CC cells, in addition to generating tumors, caused metastasis in 40% of lymph nodes, while the inoculation of only cancer cells did not generate metastasis (161). EMC proteins also contribute to the progression of CC generated by CAFs. In vitro assays showed that CC cells mainly express integrin α6β4 laminin receptors and, are conveniently capable of inducing laminin expression by CAFs, event that, together with the secretion of MMP-7 by CC cells for degradation of the basement membrane, favors the invasion of CC (162). These results show the cellular intercommunication in the TME and the essential role played by CAFs in the pro-tumor remodeling of the EMC.

Various subtypes of CAFs associated with the progression of this neoplasm have been reported; periostin-positive CAFs are involved in lymph node metastasis and with poor survival of CC patients due to the promotion of the permeability of lymphatic endothelial barriers through activation of integrin-FAK/Src-VE-cadherin signaling pathway in lymphatic endothelial cells, which decreases the expression of VE‐cadherin favoring metastatic dissemination (163). Because CAFs are the most abundant stromal cells in TME, they are a good therapeutic target; nanoparticles are a promising tool since they can be targeted with great cellular specificity, in vitro studies showed that HeLa cells and CAFs have higher intake and retention of gold nanoparticles with surface functionalized with both polyethylene glycol (PEG) and RGD, a peptide containing integrin-binding domain, compared to fibroblasts (164), suggesting that the use of nanotechnology could be helpful for treatments already used. Additionally, CAFs overexpress ER-α in tumor tissues of patients with CC, ER-α signaling promotes the expression of genes related to cell proliferation (PDGF-C and EREG), angiogenesis (VEGF-A, VEGF-C, CTGF, and ANGPT1), metastasis (MMP-1 and COL6A1) and tumor growth (FGF1), making ER-α overexpressing CAFs an important therapeutic target (165).

These findings confirm the participation of CAFs in the carcinogenesis and progression of gynecological cancers ( Table 1 and Figure 2 ), promoting that among the first ten cancers with the highest incidence and mortality worldwide are ovarian, endometrial, and cervical cancer (2).