Ovarian cancer is considered as the leading cause of death among gynaecologic cancers. This could be due to the lack of sensitive markers, precise symptoms, and effective screening methods that can help efficient treatment protocol. Therefore, this prospective study aimed to identify and enumerate CTCs in the bloodstream of ovarian cancer patients and correlate it with the clinical outcomes. To this end, we enrolled patients diagnosed with suspected ovarian malignancy who underwent exploratory laparotomy for debulking surgery. CTCs was analyzed using flow cytometry as CD105+CD24+CD117+Epcam+ phenotype. We found CTCs in 100% of patients with primary ovarian cancer while no CTCs were found in secondary ovarian cancer or in patients with either benign or borderline ovarian tumor. The highest number of CTCs was observed among malignant patients. Moreover, significant positive correlation was found between CTCs and FIGO, and RMI, and CA-125. Taken together, our data indicate that CTCs can allow to distinguish between early and late FIGO stage, presenting it as a new cellular marker for the early detection of ovarian cancer even in asymptomatic patients.
Ovarian cancer is thought to spread directly, so detection of CTCs is useful for early detection. A recent study based on the parabiosis model, in which paired mice share blood rather than lymphatic vessels, has shown that hematogenous metastasis is an important form of ovarian cancer metastasis [26].
Due to the lack of sensitive markers, precise symptoms and effective screening methods for ovarian cancer, quantitation of circulating tumor cells (CTCs) can provide information on the stage of malignancy, the onset of disease progression, and response to therapy. Previous studies tried to detect CTCs in ovarian cancer patients by using Cell Search® technology [8] and PCR analysis [27]. although these methods may identify CTCs, they are not precise and laborious. Flow cytometry-based analysis using cell antibodies against ovarian cancer can be an important alternative. So far, however, there is no efficient method has been established to detect CTCs in ovarian cancer patients. In a previous study, the use of tumor-specific fluorescent antibodies were much less efficient in quantitating CTCs than the use of tumor-specific fluorescent ligands like folate receptors to label CTCs in peripheral blood for determining the number of cancer cells circulating in the bloodstream [24]. Most recently, CTCs were identified in ovarian cancer patients by using a cocktail of antibodies including CD44, CD133, ALDH, EpCAM, and cytokeratin [25]. This study found that CD133 + ALDH + CTCs have the greatest prognostic potential in ovarian cancer among the phenotypes studied. With our aim to better quantitate CTCs more accurately, we have designed fluorescent conjugates of four specific antibodies including CD105, CD24, CD117, and Epcam by flow cytometry. We used CD117 to be able to identify CTCs with stem cell-like phenotype. To the best of our knowledge, this is the first study to analyze CTCs in ovarian cancer patients with primary and secondary and with debulking tumor.
Our ovarian cancer patients included 17 cancer patients, where 15 patients had primary ovarian cancer and 2 patients had secondary ovarian cancer. We found CTCs with CD105+CD24+CD117+Epcam+ phenotype in 100% of patients with primary, but not secondary, ovarian cancer or in patients with benign or borderline ovarian tumors. The most common marker used for CTCs is EpCAM, a epithelial marker of cancers [28]. EpCAM expression varies among different cancer types [29], and EpCAM-based CTC detection technologies are widely applied for cancers that strongly express EpCAM, such as breast and prostate cancer. Many studies have shown that CTCs in breast and prostate cancer are EpCAM-positive, and have validated their prognostic value in either early or metastatic stage cases [30, 31]. Other epithelial-derived cancer types, such as pancreatic [32], colorectal [33], and hepatocellular cancers [34], also have a considerable detection rate of EpCAM-positive CTCs. Similarly, the presence of these EpCAM-positive CTCs predicts early distant metastasis and poorer survival of patients [33, 35, 36]
CD24 is frequently overexpressed in various human cancers and is correlated with a poor prognosis. Previous study examined the functions of CD24 in human ovarian cancer cell lines and evaluated how it contributes to the molecular mechanism underlying the regeneration of cancer stem-like cells (CSCs) through the EMT mechanism in ovarian carcinoma and they demonstrated that CD24 was expressed in 70.1% of primary ovarian carcinoma tissues, which were obtained from 174 patients, and that the expression of CD24 was an independent predictor of survival in patients with ovarian cancer, also they found the expression of CD24 has been correlated with the FIGO stage, presence of peritoneal and lymph node metastasis [37].
CD105 (endoglin) antibody binds preferentially to activated endothelial cells that are induced by tumoral factors for neoangiogenesis [38] and [39]. It stains blood vessels in or around the tumor tissues [40]. Therefore, it is accepted as a more specific and sensitive marker than others to detect tumoral angiogenesis [41, 42]. Previous study determined the intratumoralneoangiogenesis of ovarian cancer patients using endoglin (CD105) and investigate the relation with prognostic factors and the found that the mean microvessel density (MVD) with CD105 was 28.78 ± 22.20, with respect to prognostic factors, CD105 was significant for both advanced stage and suboptimal cytoreduction [43].
CD117 plays an important role in cell differentiation and survival, particularly
in its impact on CSCs. In a study on non-small cell lung cancer patients, tumor cells positively expressing CD117 exhibited CSC characteristics, such as self-renewal and chemoresistance [44]. Similar characteristics are seen in CD117 positive ovarian tumor cells in which CD117 expression is related to the “stemness” of particular cancer cells [45, 46] .
The consistent expression of CD117 on all analyzed CTCs indicate to the stem-like phenotype of these cells and accordingly their aggressiveness. They would allow us in future study to sort these cells for more molecular analysis using gene array to compare its molecular signature with the primary tumor. This would confirm the stemness of these cells and its molecular features in correlation to the disease progression.
We found that the mean number of CTCs in all patients was 0.12 ± 0.11 cells/µl, where the highest number was observed among the malignant patients. Of great interest, we found a significant positive correlation between CTCs and FIGO, RMI, and CA-125. The mean value of CTCs in all patients studied was 0.12 ± 0.11 cells/µL with a minimum of 0 cells/µL and a maximum of 0.26 cells/µL. However, in the malignant patients, the minimum CTCs count was 0.02 cells/µL and the maximum CTCs count was 0.26 cells/µL. Our results agree with the findings of a previous report in which tumor-specific fluorescent ligands were used to count CTCs by flow cytometry. In this study, CTCs were detected in 18 cases out of 20 ovarian cancer patients, with an average number of CTCs of 0.2 cells/µL and a maximum number of 3.1 cells/µL [24]. While we were not able to detect CTCs in patients with secondary ovarian cancer, another study that used Cell Search® technology to detect CTCs in ovarian cancer patients, found that no CTCs in benign disease (14 patients), 17% (5/29) of patients with ovarian cancer, and 80% (4/5) of patients with secondary ovarian cancer [47], Another study used PCR to detect CTCs in ovarian cancer patients and found CTCs in 98 out of 109 ovarian cancer patients (benign or metastatic), and the average number of CTCs was 264 (range 0-1929) per 5 ml, i.e. 1.9 cells/µL [48].
To understand the role of CTCs in the early detection of ovarian cancer, we compared CTCs of early-stage patients. We found CTCs in all early-stage FIGO patients, i.e., stages I and II. We then compared CTC count with CA-125, which is the typical tumor marker. Interestingly, we found that CA-125 was negative (less than 35 U/mL) in 5 asymptomatic patients (45.5%) and was positive (more than 35 U/mL) in 6 symptomatic patients (54.5%). In contrast CA-125, we found CTCs in all asymptomatic patients (100%); 4 patients (80%) were CA-125 negative (less than 35 u\ml) and 1 patient (20%) was CA-125 positive (more than 35 U/mL). In addition, among the patients studied, there were 6 patients with stage IA or IB. Of these, 3 patients (50%) were symptomatic and 3 patients (50%) were asymptomatic. All asymptomatic patients (100%) were CTC positive; 2 patients (66.7%) were CA-125 negative, and 1 patient (33.3%) was CA 125 positive. These results are consistent with the findings of [48], who found that of 36 early-stage patients (1 and 2), 14 patients were occult, i.e., asymptomatic. In these 14 occult patients (93%; 13/14) were CTCs positive, while 86% (12/14) were CA-125 positive. A greater difference was noted in stages 1a and 1b, where the CTC-positive rate was 100% in the 7 occult patients at these stages, whereas only 57% were CA-125 positive. Taken together, our data provide a proof of concept that CTCs count in a small sample of peripheral blood is more accurate than the use the tumor marker. This concept, however, needs to be tested in a larger population of ovarian cancer patients.
Our study shows a statistically highly significant (p-value < 0.001) positive correlation (r = 0.55) between CTCs and FIGO staging in the patients studied, while no statistically significant (p-value = 0.083) positive correlation (r = 0.24) was found between RMI and FIGO staging. These results are consistent with the finding of [47, 48].
Our results show that CTCs can be used to discriminate between early and late FIGO staging at a cutoff level of > 0.82 cells/µL, with 66.7% sensitivity, 90.9% specificity, 88% PPV and 73.2% NPV (AUC = 0.65 and p-value = 0.076). Our results also show that RMI can be used to discriminate between early and late FIGO staging at a cutoff level of > 1510, with 66.7% sensitivity, 63.6% specificity, 64.7% PPV and 65.6% NPV (AUC = 0.71 & p-value = 0.0130).
Indeed,the most important factor in determining the prognosis of patients with EOC is tumor stage. The prognosis for patients with low-grade tumors at stage IA is excellent: > 90% are disease-free after 5 years. At stage II, the 5-year survival rate is 60–80% with pelvic extension of the ovarian tumor. The 5-year survival rate is significantly lower in stages III and IV. 85% of ovarian cancer cases are diagnosed at stage III, which is associated with a 5-year survival rate of 20%. Stage IV is associated with a survival rate of less than 10% and includes distant metastases. Patients with advanced FIGO stage may benefit from 3 cycles of chemotherapy before surgery [49]. Therefore, the use of CTCs alone or in combination with CA-125 can help in offering the suitable treatment protocol during the early stage of tumor before its progression into the more aggressive stage.
In a Cochrane Database Review of 11 retrospective studies, near optimal cytoreduction with < 1 cm residual disease had better overall outcomes than suboptimal cytoreduction with > 1 cm residual disease, as optimal debulking is the most important prognostic factor [50]. Our results show that there is no statistically significant difference (p-value > 0.05) between patients with optimal debulking and patients without optimal debulking in terms of the number of CTCs. This result may be due to the absence of CTCs in patients with secondary ovarian cancer (6 patients), all of whom are very advanced and in whom optimal debulking is not possible. The absence of a statistically significant relationship between CTCs and optimal debulking was also noted previously [47]. We then compared the diagnostic performance of CTCs and RMI in predicting optimal debulking. We found that CTCs can be used to discriminate between patients with optimal debulking and patients without optimal debulking at a cutoff level of < 1.9 cells/µL, with 100% sensitivity, 60% specificity, 71.4% PPV, and 100% NPV (AUC = 0.65 & p-value = 0.063). When using RMI to discriminate between patients with optimal debulking and patients without optimal debulking at a cutoff level of < 213, the results were 53.3% sensitivity, 100% specificity, 100% PPV and 68.2% NPV (AUC = 0.76 & p-value = 0.002). Taken together, our results confirm the possibility of using CTC count as a maker to distinguish between ovarian cancer patients with or without optimal debulking, providing a reliable, easy, and fast approach for evaluation of prognosis and the possibility of surgery.
In summary, our study demonstrates that flow cytometry coupled with tumor specific fluorescent antibodies can be used to quantitate CTCs in the bloodstream of ovarian cancer patients. This approach can further be validated in future prospective studies focused on detection of cancer recurrence, monitoring of disease progression, and assessment of response to therapy. Our data indicate that CTCs allow early detection of ovarian cancer even in asymptomatic patients. These cells can aid for in diagnosis of suspected ovarian mass as they differentiate between malignant and non-malignant ovarian masses and allow prediction of FIGO stage and optimum debulking, therefore prediction of prognosis. Most importantly, CTCs count allowed to distinguish between the early and late FIGO stage. Overall, our results conclude that CTCs can be used as a cellular biomarker for the early detection of ovarian cancer even in asymptomatic patients.