Ovarian cancer is the most lethal neoplasm of the female genital organs. Despite indisputable progress in the treatment of ovarian cancer, the problems of chemo-resistance and recurrent disease are the main obstacles for successful therapy. One of the main reasons for this is the presence of a specific cell population of cancer stem cells.
1. Introduction
Ovarian cancer (OC) is the most lethal tumor of the female genital tract due to aggressive behavior, late diagnosis and high recurrence potential. Most of the patients worldwide are admitted with advanced disease as the initial steps of cancer growth are usually clinically obscured. This is a reason why the 5-year survival in the whole patient population does not exceed 48% (data of American Cancer Society 2020.
https://www.cancer.org/cancer/ovarian-cancer/detection-diagnosis-staging/survival-rates.html, accessed on 20 December 2021). Moreover, ovarian cancer shows chemoresistance to standard platinum-based chemotherapy especially in advanced and recurrent cases, the fact which further influences poor survival. Ovarian cancer disease includes a heterogenous group of neoplasia: among them, about 90% are epithelial (subtypes: mucinous, serous, endometrioid and clear cells), as suggested by several and recent morphological and ultrastructural studies
[1]. Ovarian cancer is a heterogeneous disease which comprise malignant tumors of serous, mucinous, endometrial or clear cell histology. According to the differences of biological behavior and malignancy, OC has been divided into two types: type I tumors containing low-grade (LGOC) serous, mucinous and endometroid ovarian cancer with better prognosis and lower rate mortality, and type II highly malignant and rapidly progressing high-grade serous ovarian cancer (HGOC) with poor prognosis and mortality about 90% of all OC cases
[2][3][2,3]. Genetic expression profiling studies support this clinical classification, as type I tumors are associated with relative genetic stability and mutations of
PIK3CA,
PTEN,
BRAF,
KRAS and
ARID1A genes, while type II tumors possess high chromosomal instability, defective homologous recombination repair and are characterized mostly by
TP53 mutations, but also by
BRCA1,
BRCA2,
RB1 and
CTNNB1 gene mutations
[4][5][6][7][8][9][4,5,6,7,8,9]. Progenitor cells for type I OC are endometrial epithelial cells (for endometroid and clear cell tumors), tubal-peritoneal junction cells (for mucinous tumor) or fallopian epithelial cells and cortical inclusion cyst (CIC) epithelial cells (for LGOC), whereas for type II OC the progenitor cell originate from serous tubal intraepithelial cancer lesions (STIC) localized on tubal fimbriae. Early type I tumors frequently exist as so-called borderline tumors which do not show histologic signs of stromal invasion
[1][2][1,2]. Recent gene profiling studies allowed for a proposal of a new classification based on both gene expression pattern and histological structure. According to this classification ovarian cancer could be divided into five subtypes: mesenchymal, immunoreactive, proliferative, differentiated and anti-mesenchymal. Mesenchymal and proliferative tumors comprise for 28% and 20% of OC, respectively. Mesenchymal subtype show desmoplasia and mesenchymal gene expression pattern, proliferative subtype show limited inflammatory infiltration and activation of signaling pathways for stemness. Both subtypes have an unfavorable prognosis. Otherwise, immunoreactive and anti-mesenchymal subtypes which comprise 21% and 12% of OC, have a better prognosis. The immunoreactive subtype is characterized by extensive T cell tumor infiltration and T-cell receptor and toll-like receptor signaling, while the anti-mesenchymal subtype shows a genotype which is opposite to the mesenchymal type. Differentiated subtype observed in 17% of OC tumors has gene pattern resembling serous borderline tumors and intermediate prognosis
[10][11][12][10,11,12]. Extensive surgical debulking followed by platinum and taxane-based chemotherapy is a standard of care for invasive OC patients, however, extensive spread of tumor implants inside the peritoneal cavity, as well as a primary chemo-refractoriness or acquired chemoresistance of the tumor are responsible for unfavorable outcome. Recent studies suggest that a unique population of tumor cells called cancer stem cells (CSCs) are the most probable reason for cancer progression and therapy failure in OC.
2. Ovarian Cancer Stem Cells (OCSCs)—Markers
2.1. Cell Surface Markers
CSCs surface markers are not specific as they are also expressed on normal stem cells. Therefore, CSCs should also be identified by precisely defined behavior, such as spheroid formation or the reconstitution of tumors after transplantation to laboratory animals. Numerous markers have been suggested to identify CSCs, including OCSCs; however, their precise clinical significance is still unknown. Despite this, several surface cell markers identifying OCSCs isolated either from patient samples or experimental animals and cancer cell lines have been described (
Table 1).
Table 1.
Markers of OCSCs—function, correlation to clinicopathological features and their cell/tissue origin.
| Marker |
Function |
Origin of Studied Cells |
Reference |
Association to Clinicopathological Features |
Cell/Tissue Origin |
Reference |
| CD44+ |
Increased tumorigenicity, sphere-formation, cells self-renewal |
Primary EOC tumors, cell cultures |
[13][14][15][16] | [49,50,51,52] |
Number of CD44+ cells higher in early stage EOC and correlated with shorter PFS Expression correlated with high-grade, advanced (III/IV FIGO) EOC in younger (<60) patients Higher number of CD44+ cells correlated with chemoresistance and shorter DFI CD44+ correlated with Ki67 index, p53 positivity and tumor grade in HGSOC, mucinous and endometroid EOC |
EOC-isolated cells Recurrent EOC (88% HGSOC) Primary and recurrent EOC (78% HGSOC) EOC (HGSOC 62%) and BOT |
[17][18][19][20] | [105,139,140,141] |
| CD44 v6+ |
Increased tumorigenicity, recapitulation of tumors |
Xenotransplantation model |
[21] | [57] |
Distant metastases more frequent and metastasis free survival shorter in CD44v6+—high group of patients Increased number of CD44v6+ cells in primary tumors correlated with shorter OS |
EOC FIGO I–III tumors EOC FIGO III–IV tumors (71% HGSOC) |
[22][21] | [56,57] |
| CD44+/MyD88+ |
Increased tumorigenicity, sphere-formation, resistance to apoptosis, chemoresistance |
Cell lines, ascites |
[23] | [142] |
Expression of MyD88 protein was an unfavorable prognostic factor for EOC patients |
Benign ovarian tumors, BOT and EOC (54% HGSOC) |
[24] | [97] |
| CD44+/CD117+ |
Increased tumorigenicity, sphere-formation, recapitulation of tumors, chemoresistance |
EOC tumors, xenograft models |
[13] | [49] |
CD44+CD117+ cell lines were less prone to paclitaxel-induced apoptosis |
EOC cell lines |
[23] | [142] |
| CD44+/CD24- |
Increased tumorigenicity, sphere-formation |
Cell lines |
[25] | [143] |
>25% CD44+/CD24- cells in ascites correlated with higher risk of recurrence and shorter PFS |
Ascites-isolated cells from advanced EOC |
[26] | [104] |
| CD44+/CD24+/ EpCAM+ |
Increased tumorigenicity, chemoresistance |
Cell lines, EOC-isolated cell lines, ascites |
[27][28] | [100,110] |
Ovarian cancer stem cells expressing EpCAM+ are less prone to chemotherapy and are a source of recurrent tumor after the treatment |
EOC I-IV FIGO stage (45% HGSOC, 14% clear cell, 17% endometroid, 12% mucinous) |
[27] | [100] |
| CD44+/CD166+ |
Increased tumorigenicity, sphere-formation |
Cell lines |
[29] | [108] |
Population of platinum-resistant cells is enriched in CD44+/CD166+ population |
EOC-isolated and standard cell lines |
[30] | [144] |
| CD44+ALDH1+ |
Increased tumorigenicity, chemoresistance |
Cell lines |
[31] | [145] |
>50% ALDH1+ cells correlated with shorter OS |
Advanced EOC (73% HGSOC) |
[31] | [145] |
| CD44+/CD133+/ALDH1A1+ |
Chemoresistance |
Cell lines, EOC-isolated cell lines |
[32] | [116] |
Expression of markers increased in recurrent compared to primary tumors |
Advanced primary and recurrent EOC |
[32] | [116] |
| CD133+ |
Increased tumorigenicity, enhanced vasculogenesis |
Cell lines, EOC tumors, xenograft models, ascites |
[33][34][35][36] | [72,78,81,146] |
Expression of CD133+ correlated with presence of HGSOC, higher FIGO stage, ascites, chemoresistance, shorter PFS and OS No correlation with prognosis Expression of CD133+ correlated with shorter PFS and OS Expression of CD133+ correlated with shorter OS and platinum chemo-resistance |
EOC (67% HGSOC) EOC FIGO III–IV (72% HGSOC) Advanced metastatic HGSOC Advanced primary HGSOC |
[37][38][39][40] | [73,147,148,149] |
| CD133+/ALDH1A+ |
Increased tumorigenicity, cells self-renewal, chemoresistance |
EOC tumors, cell lines, xenograft models |
[41][35] | [80,81] |
Expression of CD133+ correlated with III/IV FIGO stage, expression of CD133+/ALDH1A+ correlated with shorter PFS and OS |
HGSOC |
[42] | [150] |
| CD117+ |
Increased tumorigenicity, sphere-formation, recapitulation of tumors, chemoresistance |
EOC-isolated cell lines, xenograft model, ascites |
[43][44][45][46][47] | [62,63,64,151,152] |
Expression of CD117+ correlated with shorter PFS 40% of HGSOC were CD117+ and expression correlated with chemoresistance |
Advanced metastatic HGSOC HGSOC |
[44][39] | [63,148] |
| CD24+ |
Increased tumorigenicity, stimulation of EMT |
Cell lines |
[48] | [92] |
Expression of CD24+ correlated with FIG stage and the presence of peritoneal and lymph node metastases |
27% HGSOC 12% mucinous 18% clear-cell 18% endometaroid 23% others |
[48] | [92] |