The prognostic role of the cancer stem cell marker CD44 in ovarian cancer: a meta-analysis

Ovarian cancer is the leading cause of death from gynecologic malignancy [1]. Although most ovarian cancer patients initially respond very well to platinum-based chemotherapy, the vast majority of patients will ultimately develop cancer recurrence and succumb to chemo-resistant disease [1].

CSCs represent a subpopulation of cancer cells that possess tumor initiation and self-renewal capacity and have been implicated in driving tumor growth, metastasis, and relapse following therapy in a wide variety of human cancers such as breast and ovarian cancers [2, 3]. In recent years, CSCs have also been found to contribute to the poor clinical outcome of ovarian cancer patients [4]. Furthermore, it has been suggested that CSC markers, such as CD44, ALDH1 and CD133, may serve as valuable prognostic indicators for ovarian cancer [4, 5]. Among these CSC markers, CD44 is the most frequently reported in ovarian cancer. Several studies have demonstrated that CD44 can be used to isolate cancer cells with stem cell-like and cancer-initiating properties from other populations of cancer cells [6‐8].

CD44, which consists of four functional domains, plays important roles in cell–matrix adhesion, signal transduction, cytoskeletal rearrangement, and cell migration [9]. Intriguingly, the proximal extracellular domain is the site for alternative splicing of the CD44 mRNA that results in various isoforms of CD44. The CD44 standard form (CD44s) and CD44 variants (CD44v) participate in cell–cell and cell–matrix interactions, cell migration, lymphocytehoming, and tumorigenesis. CD44v6 has particularly gained a lot of attention in recent years, since the expression of this variant has been found to be upregulated in a variety of epithelial malignancies, such as head and neck, colon, endometrium and ovarian cancers. CD44v6 may be implicated in the activation of PI3K/Akt and MAPK pathways, which can then inhibit apoptosis and promote invasion and metastasis of cancer cells [10‐14].

Although the association between the expression of CD44 or its isoforms and the survival of patients with ovarian cancer has been well investigated, the prognostic values of CD44, CD44s, and CD44v6 in predicting the survival of patients with ovarian cancer remains controversial [15‐19]. In the present study, we performed a systematic review and meta-analysis of the published literature to examine the association of the expression of CD44 or its isoforms with the clinicopathological features and the prognosis of ovarian cancer patients. These findings may help to uncover valuable marker that may enable the prognostic stratification of ovarian cancer patients and guide the management of these patients in the future.

It has been hypothesized that the formation and progression of cancers are driven by CSCs which represent a minor population in cancer cells [35]. More importantly, CSCs are considered to be responsible for chemotherapy resistance, metastasis, and postoperative recurrence [36]. A significant fraction of ovarian cancer patients will ultimately develop cancer recurrence and succumb to chemo-resistant disease [37]. Furthermore, to date, there are still no reliable markers available for the diagnosis and prognosis of ovarian cancer patients. Therefore, it remains urgent to search for a reliable prognostic parameter applicable in clinical practice to predict disease outcomes in ovarian cancer. CD44 is an important cell surface marker for isolating CSCs from tumors and is correlate with poor prognosis in various human cancers [7, 38‐40]. Intriguingly, in this study, we found that the expression of CD44v6 in ovarian cancers is significantly associated with a high TMN stage. Moreover, our findings revealed that CD44 expression is inversely correlated with a poor 5-year OS. However,we find no significant association between CD44 positivity and tumor grade, lymphatic metastasis, age of the patients, residual tumor size, response to chemotherapy, or ascites volume. These data indicate that CD44 expression may be used in the pathological evaluation of tissue histology to predict ovarian cancer prognosis in the future.

CD44 was identified as a surface glycoprotein and a lymphocyte homing receptor found on lymphoid and epithelial cells in 1982 [41]. Its main function on lymphocytes is mediating interaction with the endothelium [42]. Substantial evidence indicates that CD44 has been implicated in cancer invasion and metastasis. CD44v6, one of the major variants of CD44, could modulate the conjugation of CD44s and hyaluronic acid (HA), or enhance the metastasis of tumor by conjugating with HA [43]. CD44 plays an essential role in epithelial mesenchymal transition (EMT), one of the most important events in the cancer invasion process [44, 45]. Consistently, in some epithelial cells, the EMT process was accompanied by CD44 isoform switch from CD44v6 to CD44s, and CD44s has been proved to promote the EMT process [46]. Moreover, CD44 also plays a critical role in cell migration. After activated by its binding to hyaluronan, the cytoplasmic tail of CD44 in turn bind to the actin cytoskeleton and it would be translocated to the leading edge of the migrating cells. Thereafter, CD44 binds to CD62 on the endothelial cells, enabling the migrating cells to roll on the endothelia cells, which is the initial step of cell migration named extravasation [47]. Because of the important role that CD44 plays in cancer invasion and metastasis, it has been suggested that the expression of CD44 or certain CD44 variants could serve as valuable candidates for early detection, or as a prognostic predictor for gynecologic malignancies. Although some studies reported that high levels of CD44 expression was associated with a poor prognosis in ovarian cancer patients [17, 28], On the contrary, other groups concluded that CD44 expression had no influence on the survival of patients with ovarian cancer [15, 27]. This inconsistency may result from small sample size in each individual study. Here, we summarized the pooled data from these studies to increase the statistical power and better evaluate the prognostic values of the expression status of CD44 and its variants in ovarian cancer. To the best of our knowledge, this is the first meta-analysis of published data to evaluate the association between CD44 expression and prognosis in ovarian cancer.

It is important to note that there may be some potential limitations regarding our meta-analysis. First, variation in definitions of clinical outcomes, measurements and experimental procedures might contribute to between-study heterogeneity. It is particularly difficult to address the issue of variations in the definitions of clinical outcomes among different studies. Second, potential publication bias may also be a concern. We restricted our review to articles published in English or Chinese language because other languages were not accessible to the readers, which could favor the positive data that are more often published in English while the negative ones tend to be more often reported in other native languages.

Our findings demonstrate that CD44 expression is associated with a higher tumor TNM stage among ovarian cancer patients. Moreover, ovarian cancer patients with positive CD44 expression exhibit a worse clinical outcome than those with negative CD44 expression. Further studies with larger sample sizes will be warranted to validate the findings of our meta-analysis in the future.