Abstract
Purpose
Hepatocellular carcinoma (HCC) is one of the most common cancers and a leading cause of death worldwide. Accurate prognosis prediction tools are urgently needed. While the use of circulating tumor cells (CTCs) as prognostic prediction tool has a clear potential.
Methods
We established a comprehensive, negative enrichment-based strategy for CTCs analysis in patients with HCC, involving identification of epithelial CTCs (E-CTCs) and mesenchymal CTCs (M-CTCs) through specific biomarker. This strategy was performed in 127 HCC cases, 21 nonmalignant liver disease (NMLD) patients and 42 health control to analyze the relevance between CTCs and tumor recurrence.
Results
The total CTC number and M-CTC percent were positively correlated with tumor malignancy and high recurrence risk. Individually, preoperative total CTC number and M-CTC percent could robustly distinguish relapse cases from those with no relapse, with sensitivity of 80.95% and 90.48%, specificity of 74.12% and 84.71%, respectively. Levels of preoperative total CTC number and M-CTC percent can both be regarded as independent risk factors for HCC with early recurrence (P = 0.0053, P < 0.0001), and are both significantly correlated with worse recurrence-free survival (RFS) (log rank P < 0.0001; HR 7.78, 95% CI = 3.59–16.87; log rank P < 0.0001; HR 24.4, 95% CI = 8.67–68.77). The levels of total CTC number and M-CTC number had higher effectiveness than alpha fetal protein (AFP) in HCC longitudinal supervision (77.78% vs 88.89% vs 22.22%).
Conclusion
Preoperative and postoperative CTCs with higher effectiveness than AFP in prognosis prediction and recurrence supervision, indicating that CTCs could work as the biomarker for HCC clinical management.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Introduction
Hepatocellular carcinoma (HCC) was responsible for an estimated 905,000 new incidence and nearly 830,000 mortalities in 2020 world-wide (Sung et al. 2021). It ranks as the fourth most common malignant cancer and the third most frequent cause of cancer related death in China (Chen et al. 2016). The etiology of HCC has a certain correlation with population and geographical demographics. Approximately 80% of patients with HCC in China are associated with hepatitis B virus (HBV) infection (El-Serag and Rudolph 2007). As a crucial environmental risk factor, uncontrolled dietary aflatoxin exposure is pervasive not only in China, but also in sub-Saharan Africa and Southeast Asia (Liu and Wu 2010). Other known risk factors of HCC include hepatitis C virus (HCV) infection, diabetes, obesity, smoking, alcohol consumption, nonalcoholic fatty liver disease, hereditary diseases, and being male (Bosetti et al. 2014; Kamangar et al. 2006).
Traditional imaging methods, such as computer tomography (CT), magnetic resonance imaging (MRI) and positron emission tomography computed tomography (PET-CT) are inefficient in evaluating disease malignancy and monitoring the minimal residual disease (MRD) in some cases (Schraml et al. 2015). Currently, alpha fetal protein (AFP) is the only approved serum biomarker for HCC, however, the low sensitivity of tests for AFP limits its application (Johnson 2001). The diagnostic and prognostic efficacy of other serum protein biomarkers, such as AFP-L3 and Des-γ-carboxy (DCP), are only slightly higher than that of AFP (Wang et al. 2005; Wei et al. 2018). At present, the existing clinical staging systems, such as Okudo staging, Barcelona Clinic Liver Cancer (BCLC) staging and American Association of Cancer (AJCC) staging, have limited effect on HCC staging and prognosis estimation, and cannot be used solely based on preoperative clinical data (Huang et al. 2005; Ishizawa et al. 2008).
Therefore, there is a great need for non-invasive biomarkers which can efficiently classify the malignant degree of HCC and accurately predict the chance of postoperative recurrence. Liquid biopsy, measuring tumor byproducts such as circulating tumor cells (CTCs), cell-free tumor DNA (cfDNA) and so on, is an innovative testing method which can supplement the flaws of measuring only current serum biomarkers (Palmirotta et al. 2018). The tumor related information revealed by liquid biopsy has potential for early diagnosis, treatment response monitoring and presage tumor recurrence. As an important measured output of liquid biopsy, CTCs are tumor cells with high colonization ability and metastatic potential in the peripheral blood. There is a clear correlation between the number and phenotype of CTCs and the progression of tumor, which conducted to a valid window to understand the nature of tumor lesions (Amin et al. 2017). The concept that CTCs can be used to monitor tumor progression, metastasis, recurrence and prognosis is widely recognized (Tjensvoll et al. 2014, Alix-Panabieres and Pantel 2014).
After invading into the circulation system, CTCs can evade recognition and elimination by the immune system, then spread to distant organs and form metastatic lesions (Cho et al. 2012). The epithelial–mesenchymal transition (EMT) and its reverse process, the mesenchymal–epithelial transition (MET) are believed to play key roles in this process. EMT is a complicated process that endows epithelial cells with mesenchymal cell traits and enhanced metastatic and invasive capability (Vardas et al. 2021). During EMT, the expression of epithelial markers, such as the epithelial cell adhesion molecule (EpCAM) and cytokeratins (CK), will be downregulated, while expression of mesenchymal markers, such as vimentin and twist, will be upregulated (Kalluri 2009). The change between epithelial and mesenchymal states is plastic; CTCs may go through dynamic processes of EMT and MET, and acquire various phenotypes on the spectrum of epithelial to mesenchymal. CTCs with mesenchymal markers have been found in multiple cancers, including breast, lung, prostate, and liver cancer (Khoshbakht et al. 2021; Jiang et al. 2021).
The combination of the epithelial and mesenchymal markers has potential as a relatively more sensitive and accurate CTCs detection strategy compared to current methods. According to the previous researches, vimentin can serve as an identification biomarker of CTCs undergoing EMT (Satelli et al. 2015; Li et al. 2018). To detect both the epithelial and mesenchymal CTCs in peripheral blood of HCC patients, the antibody combination of CK, vimentin and CD45 was applied in this study. After excluding white blood cells (WBC) according to the expression of CD45, we classified the CTCs based on expression of vimentin and CK as either epithelial CTCs (E-CTCs), mesenchymal CTCs (M-CTCs) and epithelial–mesenchymal CTCs (EM-CTCs). We then investigated the correlation between clinical/pathological features and different phenotype of CTCs, and evaluated its potential prognostic value. It turned out that the M-CTC percent was a more effective indicator than total CTC number in HCC recurrence prediction. Also, total CTC number > 10/7.5 ml and M-CTC percent of > 23.33% were correlated with a worse recurrence free survival (RFS) rate, also could be used to independently predict HCC early recurrence. In addition, CTCs could work as an effective recurrence indicator in HCC supervision, especially for the AFP negative patients.
Methods
Patient enrollment and follow-up
From March 2017 to September 2019, a total of 127 HCC patients from the Cancer Hospital of Chinese Academy of Medical Sciences were included in the study group. All patients were diagnosed with HCC by pathologic and fine-needle aspiration biopsy. Another 42 healthy controls and 21 patients with nonmalignant liver disease (NMLD) (adenoma: N = 1; liver cirrhosis without HCC: N = 15; focal nodular hyperplasia: N = 2; hepatic hemangioma: N = 3) were enrolled. The tumor stages of HCC patients were defined according to the Barcelona Clinic Liver Cancer (BCLC) and American Joint Committee on Cancer (AJCC) staging systems. Tumor differentiation was stratified based on the Edmondson grading system. Clinical information such as age, sex, hepatitis virus infection status and serum tumor biomarkers were collected. In addition, the biochemical examination, imaging data and pathological results of HCC patients were collected. The clinical characteristics of HCC patients, healthy controls and NMLD patients were presented in Table S1, Table S2 and Table S3, respectively. After operation, patients were followed up every 3 months during the first postoperative year, and every 6 months afterwards. Patients suspected to have recurrence and metastasis were monitored by serum AFP, abdomen ultrasonography, CT, MRI and/or PET CT. Follow-up was terminated on September 30, 2021, and recurrence-free survival (RFS) was defined as the time interval between the operation and recurrence or the endpoint of follow-up. The study was approved by the ethics review committee of Cancer Hospital of Chinese Academy of Medical Sciences (ID: NCC3094). Informed consent was obtained from participants in accordance with respective committee regulations and the study was performed in accordance with the guidelines of the Declaration of Helsinki.
Establishment of recurrence risk index
A high recurrence risk index was established based on methods in previous studies characterizing the pathological character of tumors. The criteria included: tumor number > 1; tumor diameter > 5 cm; macroscopic vascular invasion (tumor thrombus); pathologic microvascular invasion (MVI); satellite nodules; the T stages of AJCC was 2, 3 or 4; histopathological diagnosis was poorly differentiated or undifferentiated, the index over 0 was defined as high recurrence risk patient (Xu et al. 2019; Shah et al. 2007).
Blood sample collection and CTCs enrichment
The first 2 ml blood was discarded to avoid contamination from epithelial cells of the skin. Then, 8–10 mL blood sample was collected in Streck tubes (STRECK, USA) and transferred to laboratory. Then, 7.5 ml blood sample was precisely added into centrifuge tube with pipette, then it was diluted with equal volume of phosphate buffered saline (PBS) buffer under the aseptic condition. 3 ml of Histopaque-1077 reagent (Sigma–Aldrich, USA) added into the bottom of Leucosep® tube (Greiner Bio-One, Germany) and balanced to room temperature. Then, the diluted blood sample was added into the upper channel of Leucosep® tube and centrifuged at 1000 g/10 min at room temperature to separate peripheral blood mononuclear cells (PBMC). The buffer coat containing the PBMC was transferred to a new 15 mL tube, and cells were gently resuspended with 10 mL washing buffer (0.5% Bovine Serum Albumin (BSA), 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 2 mM KH2PO4, 1 mM EDTA, pH 7.4), and centrifuge at 300g at 4 °C for 10 min, the supernatant was discarded and this was repeated one more time to obtain nucleated cells. Then, anti-human CD45 immunomagnetic beads (Miltenyi Biotec, Germany) were added in equal proportion (20 ul/107 WBC) and incubated at 4 °C for 15 min. The supernatant was removed, 500 uL buffer was added to resuspend the pellet, then the cell suspension was transferred to the LS column (Miltenyi Biotec, Germany) under a strong magnetic field, eluted and nucleated cells not bound to the CD45 magnetic beads (including CTCs) were collected. The supernatant was transferred to a new tube centrifuged at 1000g/10 min and subsequently spotted on a glass slide.
CTCs identification
The CTC droplet slides were blocked with PBS containing 2% BSA for 30 min, and then anti Vimentin-AlexaFluor 647 (1:200) (Cell Signaling Technology, CAT: 98565), anti-CK8/18/19-FITC (1:100) (Miltenyi Biotec, CAT: 130-080-101) and anti-CD45- PE (1:200) (Miltenyi Biotec, CAT: 130-045-801) diluted in 2% BSA were added and incubated at room temperature for 1 h. Then, the slides were washed three times with PBS, 3 min/time to remove unbound antibody. The CTC slides were mounted with the nuclear dye 4′,6-diamidino-2-phenylindole (DAPI) (Sigma–Aldrich, CAT: D9564) and observed under a fluorescence microscope (Nikon, Japan). The epithelial CTCs (E-CTCs) were defined as PanCK (+), vimentin (−), CD45 (−) and DAPI (+) cells, the mesenchymal CTCs (M-CTCs) were defined as PanCK (−), vimentin (+), CD45 (−) and DAPI (+) cells, the epithelial–mesenchymal CTCs (EM-CTCs) were defined as PanCK (+), vimentin(+), CD45 (−) and DAPI (+) cells, whereas white blood cells (WBCs) were defined as PanCK (−), vimentin (−), CD45 (+) and DAPI (+) cells, respectively. Total CTC number were calculated as the sum of E-CTCs, M-CTCs and EM-CTCs of each patient. For some patients whose blood collection does not reach 7.5 ml, the final number of CTC is calculated by the following formula: number of CTC/volume of blood used in the experiment * 7.5 ml.
Statistical analysis
The statistical analysis was conducted through GraphPad Prism version 9.0 software (GraphPad Software, USA) and R software (Version 3.6.1, R Foundation for Statistical computing, Austria). All measured data are presented as the mean ± standard deviation (mean ± SD). The Mann–Whitney rank test was used to compare the differences between two groups, and the analysis of variance (ANOVA) was used to compare the differences among the more than two groups by multiple comparisons test. The “pROC” package of R was performed to establish the receiver operating characteristic (ROC) curve to evaluate the diagnostic and prognostic value of different subtype of CTCs. The “survival” and “survminer” were used to conduct univariate and multivariate analysis of Cox regression model. The continuous parameters and the total CTC number and M-CTC percent were dichotomized for RFS before the log-rank test by optimal cutoff values determined based on the ROC curve. All statistical tests were two-tailed and considered significant if P < 0.05 (*P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001).
Results
Patient characteristics
A total of 127 HCC patients were enrolled in this study. The clinical and pathological characteristics of the 127 recruited patients are summarized in Table S1. The cohort includes 106 male patients (83.46%) and 21 female patients (16.53%), with a median age of 55 (range 28–77). 39 patients (30.71%) had received adjuvant therapy before hepatectomy. 87.40% (111/127) of patients were hepatitis B virus (HBV) positive and 8.66% (11/127) were hepatitis C virus (HCV) positive. In addition, 116 patients (91.34%) presented with cirrhosis. The cohort was largely composed of early stage HCC patients, with 66.93% (85/127), 11.02% (14/127) and 22.05% (28/127) of patients in BCLC stage A, B and C, respectively, and 42.06% (53/127), 36.51% (46/127), 20.63% (26/127) and 0.79% (1/127), in AJCC stages I, II, III, and IV, respectively (Table S1). In addition, 42 healthy control subjects, including 29 males (69.05%) and 13 females (30.95%) with a median age of 54 (range 23–76), and 21 patients with nonmalignant liver disease (NMLD), including 18 males (85.71%) and 3 females (14.29%) with a median age of 52 (range 32–72), were recruited at the same time. The clinical characteristics are summarized in Tables S2 and S3, respectively.
Identification of different EMT phenotypes of CTCs
EMT-related CTCs in the preoperative peripheral blood of 127 HCC patients after negative enrichment were identified using Pan-CK, vimentin and CD45 antibodies. Sample images of E-CTCs, M-CTCs, EM-CTCs and a few residual CD45+ white blood cells (WBC) are shown in Fig. 1a–c. In all HCC patients, the detection rate of E-CTCs (92.91%, 118/127) was slightly higher than that of M-CTCs (79.53%, 101/127) and EM-CTCs (81.89%, 104/127). The median numbers of E-CTCs, M-CTCs and EM-CTCs in 7.5 mL of preoperative peripheral blood from HCC patients in our cohort were 5 (range 0–20), 2 (range 0–14) and 2 (range 0–8), respectively.
Correlation between tumor malignancy and different subtypes of EMT-related CTCs
A high recurrence risk index was established based on previous studies, using pathological characteristics of the tumor. A patient with an index over 0 was defined as a high recurrence risk patient (Xu et al. 2019; Shah et al. 2007). Among recruited HCC patients, 34.65% (44/127) were low risk (high recurrence risk index = 0), and 64.57% (82/127) were high risk (high recurrence risk index > 0). In the healthy control group, NMLD group, low risk, and high risk group, the median numbers of total CTCs (the sum of E-CTCs, M-CTCs and EM-CTCs) were 0/7.5 ml (range 0–2.5/7.5 ml), 0/7.5 ml (range 0–8/7.5 ml), 8/7.5 ml (range 0–20/7.5 ml) and 12/7.5 ml (range 0–34/7.5 ml), respectively (Fig. 1d). More total CTCs were found in HCC patients compared to healthy controls or NMLD patients (low risk: P < 0.0001, P = 0.0002; high risk: P < 0.0001, P < 0.0001), meanwhile, more CTCs were found in high risk HCC patients compare with low risk HCC patients (P < 0.0001) (Fig. 1e). Similarly, the percent of M-CTC was found much higher in HCC patients compared to NMLD patients or healthy controls (low risk: P < 0.0001, P = 0.0039; high risk: P < 0.0001, P < 0.0001), the percent of M-CTC in high risk HCC patients were much higher than that in low risk HCC patients (P = 0.0003) (Fig. 1f).
Next, we evaluated the correlation between the levels of total CTC number and M-CTC percent with clinical parameters related to the poor prognosis of HCC. In this analysis, the amount of total CTCs in the peripheral blood was associated with the tumor number (P = 0.0233), tumor size as measured by the largest diameter (P = 0.0115), thrombosis (P = 0.0035), microvascular invasion (MVI, P = 0.0076), AJCC stage (P = 0.0003), and BCLC stage (P = 0.0007), indicating that total CTC levels are associated with HCC tumor burden and could help assess HCC progression (Fig. 2, Table 1).
Similarly, higher M-CTC percent was found in patients with multi tumor nodus (P = 0.0099), larger tumor sizes (P = 0.0011), thrombosis present (P = 0.0089), MVI present (P = 0.0231), higher AJCC stage (P = 0.0016), which suggests that relative M-CTC levels can assess disease progression of HCC patients in addition to total CTC number (Fig. 2, Table 1). In addition, neither total CTC number nor M-CTC percent were correlated with other clinicopathological features (sex: P = 0.6020, P = 0.5159; age: P = 0.4691, P = 0.8054; HBV infection: P = 0.0688, P = 0.2300; HCV infection: P = 0.2527, P = 0.7678; differentiation: P = 0.5986, P = 0.0710) (Table 1).
M-CTCs could work as a more precise prognostic biomarkers in HCC patients
Based on stratification of the high risk index, we demonstrated that total CTC number and M-CTC percent could be used to evaluate recurrence risk of HCC patients (Fig. 1e–f). Thus, the recurrence prediction ability of total CTC number and M-CTC percent were explored. Among the 127 HCC patients, the median follow-up time was 38.62 months after hepatectomy (range 1.43–51.13 months), and 42 patients (33.07%) experienced recurrence or distant metastasis as confirmed by imaging test. Total CTC number and M-CTC percent were higher in post-recurrence patients compared to non-recurrence patients (P < 0.0001; P < 0.0001) (Fig. 2f, Table 1). Healthy controls and NMLD patients were also followed for a comparable time, and none developed HCC.
ROC curve analysis was then applied to evaluate the ability of different EMT-related subtype of CTCs and serum tumor biomarker (AFP and CEA) to predict early recurrence/metastasis of HCC. Compared with E-CTC number (AUC = 0.724), the traditional identification subtype of CTCs, the total CTC number (AUC = 0.801) and M-CTC number (AUC = 0.873) both performed better in recurrence prediction, much higher than the AFP (AUC = 0.543) and CEA (AUC = 0.528) (Fig. 3a). Unsurprisingly, the M-CTC percent slightly improved the prognostic ability with an AUC of 0.882 (Fig. 3b). The cut off value used for total CTC number and M-CTC percent were 10/7.5 ml and 23.33%, respectively, which were selected based on the Youden index. Based on this cut off value, both total CTC number and M-CTC percent individually could distinguish 34/42 and 38/42 relapse cases through 2 years follow-up. Additionally, total CTC number and M-CTC percent accurately predicted 63 and 72 of 85 HCC cases without relapse or metastasis. Thus, measuring total CTC number and M-CTC percent yielded sensitivity rates of 80.95% and 90.48%, and specificity rates of 74.12% and 84.71%, respectively, which were significantly higher than when measuring AFP (cut off = 20 ng/ml, sensitivity 59.52%, specificity 44.71%) and CEA (cut off = 5 ng/ml, sensitivity 16.67%, specificity 88.24%) (Fig. 3c, d). Altogether, total CTC number and M-CTC percent had rigid estimate values that were significantly better than those of serum protein AFP and CEA.
Correlation between the total CTC number and M-CTC percent and the prognosis of HCC patients
Next, Kaplan–Meier analysis was performed to assess the ability of total CTC number and M-CTC percent to predict risk of recurrence. Positive levels of total CTC number (> 10/7.5 ml) and M-CTC percent (> 23.33%) determined by the ROC curve were applied. 34 of the 56 (60.71%) patients positively determined by total CTC number did go through cancer recurrence, while only 8 of the 71 (11.27%) total CTC number-negative patients went through recurrence (Fisher’s exact test, P < 0.0001). Similarly, 38 of the 51 (74.51%) M-CTC percent-positive patients experienced HCC recurrence, while only 4 of the 76 (5.26%) M-CTC percent-negative patients (Fisher’s exact test, P < 0.0001) had recurrence. The RFS of preoperative total CTC number-positive patients was significantly lower than in total CTC number-negative patients (median RFS for positive and negative groups, 15.5 months vs not reached, log rank P < 0.0001; HR 7.78, 95% CI = 3.59–16.87) (Fig. 3e, Table 2). Similarly, M-CTC percent level was significantly correlated with worse RFS (12.6 months vs not reached, log rank P < 0.0001; HR 24.4, 95% CI = 8.67–68.77) (Fig. 3f, Table 2).
Furthermore, univariate and multivariate Cox regression analysis were conducted to investigate the independent risk factors of HCC early recurrence. Based on univariate Cox regression, presence of multiple nodules, MVI, thrombosis, AJCC stage, total CTC number and M-CTC percent were selected for multivariate Cox regression. The analysis revealed that multiple nodules (HR 3.27; 95% CI, 1.47–7.28; P = 0.0036), total CTC number (HR 3.49; 95% CI 1.45–8.40; P = 0.0053), M-CTC percent (HR 16.35; 95% CI 5.61–47.67; P < 0.0001) were independent risk factor for early HCC recurrence (Table 2). These results demonstrated that both total CTC number and M-CTC percent can function as indicating markers of early recurrence of HCC, especially when combined with pathological features, such as multiple nodules.
CTCs played indispensable role in disease surveillance and relapse alert
During follow-up, 52 HCC patients in our cohort were able for the post-operation observation and 18 of them (34.62%) went through the recurrence. We found that reduction in total CTC number and AFP levels from preoperative baseline to the first post-operation timepoint (~ 3 month after surgery) was significantly correlated with the well effect of surgical treatment (P < 0.0001 and P < 0.0001), while the reverse was occurred in relapse candidates (Fig. 4a–c), especially in M-CTC number. Among patients without progression, 100% (34 of 34) experienced decrease or stable in total CTC number, 100% (34 of 34) in M-CTC number and 91.18% (31 of 34) in AFP from preoperative baseline, while in recurrence patients, 38.89% (7 of 18) saw an ascent in total CTC number from baseline, 61.11% (11 of 18) in M-CTC number from baseline, which was conspicuous higher than 16.67% (3 of 18) of AFP. This performance indicated that M-CTCs, was a more sensitive circuiting tumor marker during longitudinal supervision.
In disease supervision of 15 HCC patients, compared with the basically stable state of 7 non-recurrence cases (Fig. 4d), the changes of CTC number in the recurrence patients were more dramatic. To be specific, total CTC number could alert 7 of 9 (77.78%) recurrence patients. The median interval indicated by CTC number > 10/7.5 ml and recurrence confirmed by MRI/CT/PET image detections was 266 days (range 98–373 days). While M-CTC number (> 2/7.5 ml) could alert 8 of 9 (88.89%) recurrence patients with median interval of 190 days (range 98–373 days). Both were more effective than AFP (> 20 ng/ml), which only valid in 2 of 9 (22.22%) patients with median interval of 322 days (range 297–347 days) (Fig. 4e). For five AFP-negative cases, the dynamic monitoring of CTCs could complement four of them with effective recurrence warning. The results above emphasized indispensability of CTCs, especially M-CTCs dynamic monitoring in early warning of HCC recurrence.
Discussion
The limitations of measuring serum protein markers in patients with HCC include low sensitivity for early diagnosis, poor prognosis prediction and inability to accurately predict recurrence. AFP is the only serum biomarker currently measured in the clinical management of patients with HCC, but has a 30% false negative rate in HCC diagnosis and even worse performance in relapse prediction (Farinati et al. 2006; Mehta et al. 2017). To address these issues, this study identification the CTCs with different EMT phenotypes: epithelial CTCs (E-CTCs), mesenchymal CTCs (M-CTCs) and CTCs with both epithelial and mesenchymal markers (EM-CTCs) in HCC patients. Then the correlation between tumor malignancy and different subtypes of EMT-related CTCs were explored, and found that total CTC number and M-CTC percent were positively correlated with tumor malignancy parameter. In addition, consistent with previous studies, preoperative total CTC number and M-CTC performed better than that of AFP in HCC recurrence prediction (Court et al. 2018; Qi et al. 2018). As for the longitudinal supervision of HCC patients, we found that CTCs, especially M-CTCs were indispensable of in early warning of HCC recurrence.
Due to their low abundance in the peripheral blood, the main technical difficulty of CTC-analysis is the lack of an effective enrichment and identification system. Previous detection methods were largely based on expression of epithelial markers, such as EpCAM and CK, which are widely expressed in solid tumors of epithelial origin. For example, the CellSearch system approved by the Food and Drug Administration for CTCs detection in the identification of metastatic breast cancer, prostate cancer and colon cancer is based on the EpCAM expression (Riethdorf et al. 2007). While regarded as the stem-like cancer cell, EpCAM-positive cells account for only 30% of HCC, thus making the CellSearch system unsuitable for CTCs detection in HCC (Ruck et al. 2000; Shen et al. 2018). In this study, our negative enrichment strategy was based on magnetic activated cell sorting (MACS) to remove the leukocytes, enriching for CTCs in the peripheral blood of HCC. The subsequent phenotype identification was performed using immunofluorescence staining of different markers. This strategy has good potential for application in the clinic due to ease of operation and minimal dependency on difficult-to-access equipment.
CTCs are highly heterogeneous, EMT-like CTCs have been demonstrated to be the possible contributors to distant metastases and tumor progression in solid tumors (Dong et al. 2020; Gao et al. 2021). EMT gives primary tumor cells the ability to break through the extracellular matrix binding, invade into the circulatory system and form distant metastases (Shook and Keller 2003). During this process, expression of epithelial markers such as EpCAM and CK decrease, while expression of mesenchymal markers rise, leading to false negatives in CTCs identification (Mikolajczyk et al. 2011). CTCs undergoing EMT share characteristics with cancer stem cells, including stronger invasive, metastasic abilities, higher tolerance to chemotherapy and targeted therapeutic drugs (Mitra et al. 2015). However, due to the lack of specific marker molecules and detection methods for EMT-like CTCs, relevant studies are limited.
Vimentin is an important component of intermediate filaments that is extensively expressed in mesenchymal cells. It is also overexpressed in breast cancer, cancers of the central nervous system, prostate cancer, malignant melanoma and lung cancer and is measured in the detection step of the CellSearch system (Satelli and Li 2011). Recent studies have found that glioblastoma cells expressing vimentin have strong proliferative ability in vitro (Noh et al. 2016). Therefore, vimentin could also be used as a broad-spectrum marker for CTC-detection, as patients with positive vimentin have worse response to treatment (Satelli et al. 2015, 2014).
In this study, vimentin was used to detect M-CTCs, and the epithelial marker CK was used to detect E-CTCs. Using these methods, we found a high heterogeneity of CTCs in the peripheral blood of patients with HCC. Furthermore, we evaluated the efficacy of the total CTC number and M-CTC percent in prognosis prediction in a large HCC cohort, and found that these markers outperformed than the serum marker AFP for early diagnosis and prognosis prediction. M-CTC percent were more effective in prognosis progression compared with total CTC number, suggesting that vimentin positive CTCs may contribute to worse outcomes. In addition, high CTC number and M-CTC percent were significantly associated with poor RFS rate in HCC patients and worked as independently significant risk factors for HCC early recurrence. We also evaluated the supervision ability of total CTC and M-CTC in longitudinal HCC patients, both worked more effectively than AFP in warning rate and lead time. Inevitably, there are some shortcomings of our research: (1) only EMT-related subtypes of CTCs were measured, which may neglect the complexity given by the full heterogeneity of CTCs; (2) The number of cases used for survival analysis is relatively small, and related conclusions require further verification in a large independent cohort. In brief, our study supports the application of CTCs as more effective and accurate than those serum biomarkers currently used in the clinic in HCC prognosis prediction and supervision, which provide a foundation for future studies.
Data availability
Data are available in a public repository. All data relevant to the study are included in the article or uploaded as online supplemental information. Additional data related to this paper may be requested from the corresponding authors.
References
Alix-Panabieres C, Pantel K (2014) Challenges in circulating tumour cell research. Nat Rev Cancer 14(9):623–631
Amin MB, Greene FL, Edge SB et al (2017) The eighth edition AJCC cancer staging manual: continuing to build a bridge from a population-based to a more “personalized” approach to cancer staging. CA Cancer J Clin 67(2):93–99
Bosetti C, Turati F, La Vecchia C (2014) Hepatocellular carcinoma epidemiology. Best Pract Res Clin Gastroenterol 28(5):753–770
Chen W, Zheng R, Baade PD et al (2016) Cancer statistics in China, 2015. CA Cancer J Clin 66(2):115–132
Cho EH, Wendel M, Luttgen M et al (2012) Characterization of circulating tumor cell aggregates identified in patients with epithelial tumors. Phys Biol 9(1):016001
Court CM, Hou S, Winograd P et al (2018) A novel multimarker assay for the phenotypic profiling of circulating tumor cells in hepatocellular carcinoma. Liver Transpl 24(7):946–960
Dong X, Ma Y, Zhao X et al (2020) Spatial heterogeneity in epithelial to mesenchymal transition properties of circulating tumor cells associated with distant recurrence in pancreatic cancer patients. Ann Transl Med 8(11):676
El-Serag HB, Rudolph KL (2007) Hepatocellular carcinoma: epidemiology and molecular carcinogenesis. Gastroenterology 132(7):2557–2576
Farinati F, Marino D, De Giorgio M et al (2006) Diagnostic and prognostic role of alpha-fetoprotein in hepatocellular carcinoma: both or neither? Am J Gastroenterol 101(3):524–532
Gao Y, Fan WH, Song Z et al (2021) Comparison of circulating tumor cell (CTC) detection rates with epithelial cell adhesion molecule (EpCAM) and cell surface vimentin (CSV) antibodies in different solid tumors: a retrospective study. PeerJ 9:e10777
Huang YH, Chen CH, Chang TT et al (2005) Evaluation of predictive value of CLIP, Okuda, TNM and JIS staging systems for hepatocellular carcinoma patients undergoing surgery. J Gastroenterol Hepatol 20(5):765–771
Ishizawa T, Hasegawa K, Aoki T et al (2008) Neither multiple tumors nor portal hypertension are surgical contraindications for hepatocellular carcinoma. Gastroenterology 134(7):1908–1916
Jiang SS, Mao CG, Feng YG et al (2021) Circulating tumor cells with epithelial-mesenchymal transition markers as potential biomarkers for the diagnosis of lung cancer. World J Clin Cases 9(12):2721–2730
Johnson PJ (2001) The role of serum alpha-fetoprotein estimation in the diagnosis and management of hepatocellular carcinoma. Clin Liver Dis 5(1):145–159
Kalluri R (2009) EMT: when epithelial cells decide to become mesenchymal-like cells. J Clin Invest 119(6):1417–1419
Kamangar F, Dores GM, Anderson WF (2006) Patterns of cancer incidence, mortality, and prevalence across five continents: defining priorities to reduce cancer disparities in different geographic regions of the world. J Clin Oncol 24(14):2137–2150
Khoshbakht S, Azimzadeh Jamalkandi S, Masudi-Nejad A (2021) Involvement of immune system and epithelial-mesenchymal-transition in increased invasiveness of clustered circulatory tumor cells in breast cancer. BMC Med Genomics 14(1):273
Li H, Meng QH, Noh H et al (2018) Cell-surface vimentin-positive macrophage-like circulating tumor cells as a novel biomarker of metastatic gastrointestinal stromal tumors. Oncoimmunology 7(5):e1420450
Liu Y, Wu F (2010) Global burden of aflatoxin-induced hepatocellular carcinoma: a risk assessment. Environ Health Perspect 118(6):818–824
Mehta NJ, Celik AD, Peters MG (2017) Screening for hepatocellular carcinoma: what is missing? Hepatol Commun 1(1):18–22
Mikolajczyk SD, Millar LS, Tsinberg P et al (2011) Detection of EpCAM-negative and cytokeratin-negative circulating tumor cells in peripheral blood. J Oncol 2011:252361
Mitra A, Satelli A, Xia X et al (2015) Cell-surface Vimentin: a mislocalized protein for isolating csVimentin(+) CD133(-) novel stem-like hepatocellular carcinoma cells expressing EMT markers. Int J Cancer 137(2):491–496
Noh H, Yan J, Hong S et al (2016) Discovery of cell surface vimentin targeting mAb for direct disruption of GBM tumor initiating cells. Oncotarget 7(44):72021–72032
Palmirotta R, Lovero D, Cafforio P et al (2018) Liquid biopsy of cancer: a multimodal diagnostic tool in clinical oncology. Ther Adv Med Oncol 10:1758835918794630
Qi LN, Xiang BD, Wu FX et al (2018) Circulating Tumor cells undergoing EMT provide a metric for diagnosis and prognosis of patients with hepatocellular carcinoma. Cancer Res 78(16):4731–4744
Riethdorf S, Fritsche H, Muller V et al (2007) Detection of circulating tumor cells in peripheral blood of patients with metastatic breast cancer: a validation study of the Cell Search system. Clin Cancer Res 13(3):920–928
Ruck P, Wichert G, Handgretinger R et al (2000) Ep-CAM in malignant liver tumours. J Pathol 191(1):102–103
Satelli A, Li S (2011) Vimentin in cancer and its potential as a molecular target for cancer therapy. Cell Mol Life Sci 68(18):3033–3046
Satelli A, Mitra A, Cutrera JJ et al (2014) Universal marker and detection tool for human sarcoma circulating tumor cells. Cancer Res 74(6):1645–1650
Satelli A, Brownlee Z, Mitra A et al (2015) Circulating tumor cell enumeration with a combination of epithelial cell adhesion molecule- and cell-surface vimentin-based methods for monitoring breast cancer therapeutic response. Clin Chem 61(1):259–266
Schraml C, Kaufmann S, Rempp H et al (2015) Imaging of HCC-current state of the art. Diagnostics (basel) 5(4):513–545
Shah SA, Cleary SP, Wei AC et al (2007) Recurrence after liver resection for hepatocellular carcinoma: risk factors, treatment, and outcomes. Surgery 141(3):330–339
Shen J, Wang WS, Zhu XL et al (2018) High epithelial cell adhesion molecule-positive circulating tumor cell count predicts poor survival of patients with unresectable hepatocellular carcinoma treated with transcatheter arterial chemoembolization. J Vasc Interv Radiol 29(12):1678–1684
Shook D, Keller R (2003) Mechanisms, mechanics and function of epithelial-mesenchymal transitions in early development. Mech Dev 120(11):1351–1383
Sung H, Ferlay J, Siegel RL et al (2021) Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 71(3):209–249
Tjensvoll K, Nordgard O, Smaaland R (2014) Circulating tumor cells in pancreatic cancer patients: methods of detection and clinical implications. Int J Cancer 134(1):1–8
Vardas V, Politaki E, Pantazaka E et al (2021) Epithelial-to-mesenchymal transition of tumor cells: cancer progression and metastasis. Int J Dev Biol 66:277–283
Wang CS, Lin CL, Lee HC et al (2005) Usefulness of serum des-gamma-carboxy prothrombin in detection of hepatocellular carcinoma. World J Gastroenterol 11(39):6115–6119
Wei T, Zhang W, Tan Q et al (2018) Electrochemical assay of the alpha fetoprotein-L3 isoform ratio to improve the diagnostic accuracy of hepatocellular carcinoma. Anal Chem 90(21):13051–13058
Xu XF, Xing H, Han J et al (2019) Risk factors, patterns, and outcomes of late recurrence after liver resection for hepatocellular carcinoma: a multicenter study from China. JAMA Surg 154(3):209–217
Funding
This work was partly supported by the National Natural Science Foundation of China (81930068 and 81772299 to ZW), Beijing Natural Science Foundation (7191007 to ZW) and the CAMS Innovation Fund for Medical Sciences (2021-I2M-1-052 to ZW).
Author information
Authors and Affiliations
Contributions
All authors contributed to the study conception and design. LZ: performed experiments, obtained and analyzed data. LZ and ZZ: wrote the manuscript; LZ and FL: helped with patient follow-up and clinical data interpretation; LZ and ZZ: did data analysis; ZW: conceived, designed and coordinated the experiments. ZW: guided the work, analyzed data and wrote the manuscript.
Corresponding author
Ethics declarations
Conflict of interest
All authors have completed the ICMJE uniform disclosure form. The authors have no conflicts of interest to declare.
Ethics approval
This study was carried out with the approval of the Institutional Review Board of the Ethics Committee of Cancer Hospital of Chinese Academy of Medical Sciences (ID: NCC3094). The project was approved on October 12, 2016 and reviewed annually. The study was performed in accordance with the guidelines of the Declaration of Helsinki.
Informed consent
Informed consent was obtained from all participants included in our study.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Zhao, L., Zheng, Z., Liu, Y. et al. The mesenchymal circulating tumor cells as biomarker for prognosis prediction and supervision in hepatocellular carcinoma. J Cancer Res Clin Oncol 149, 6035–6048 (2023). https://doi.org/10.1007/s00432-022-04526-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00432-022-04526-9