Abstract
Background/Aim: Clinically, some cancer patients develop drug resistance after receiving a few courses of chemotherapy, or even worse, completely lack therapeutic response. Prediction of treatment response before administration is of value to oncologists. This study aimed to evaluate the feasibility of drug sensitivity tests for circulating tumor cells (CTCs) isolated from patients with advanced hepatocellular carcinoma (HCC). Materials and Methods: CTCs isolated from patients receiving cytotoxic chemotherapy or sorafenib were subjected to drug tests using ex vivo culture. Thirty-one patients with advanced HCC and one with benign lesions were enrolled in the study. Results: After incubation with chemotherapeutic drugs ex vivo, the numbers of CTCs were decreased in 10/12 (83.3%) of treatment-naïve patients (planning to receive the first course of chemotherapy) but increased in all patients (6/6) who had received chemotherapy (p=0.002). The CTC count was negatively correlated with the overall survival of patients (p=0.016). The CTCs of patients who received targeted therapy (n=11), were incubated with sorafenib for sensitivity tests. After comparing the chemotherapy and sorafenib-treated groups, the CTCs in the latter group had a lower probability to develop drug resistance (p=0.031). Conclusion: An ex vivo culture-based drug sensitivity test was developed for CTCs isolated from advanced HCC patients. The drug test found that resistance developed rapidly following cytotoxic chemotherapy, whereas it was rarely observed in patients receiving sorafenib. For patients with advanced HCC who choose to receive chemotherapy, CTC drug sensitivity tests may help predict treatment response.
Hepatocellular carcinoma (HCC) is the fifth most common solid cancer and the third most common cause of cancer-related death worldwide (1). Chronic hepatitis B virus (HBV) infection, chronic hepatitis C virus (HCV) infection, and alcoholic liver diseases are the top three risk factors for HCC (2, 3). The processes of hepatocarcinogenesis involves not only multiple molecular events but also heterogeneous cellular pathways (4-6). Systemic therapy, including cytotoxic chemotherapy and targeted therapy (TT), such as sorafenib, has been widely used in the past decades, owing to its proven beneficial effect in prolonging patient survival. It is now the recommended first-line or second-line (next to immunotherapy) treatment according to international guidelines. However, patients with advanced HCC usually have a poor response to TT. The response rates range from 2.3 to 24% for the first-line TT and from 2.2 to 11% for the second-line TT (7). Although several studies have reported that hepatic arterial infusion chemotherapy (HAIC), such as 5-fluorouracil plus cisplatin, had a higher response rate and better overall survival than TT, there is no established standard protocol or indication for HAIC or TT in advanced HCC (8-11). Additionally, large-scale randomized control trials have not demonstrated the efficacy of cytotoxic chemotherapy, and a small percentage of patients still achieve complete or partial responses. Besides, cytotoxic chemotherapy remains an ideal option in developing or under-developed countries where multi-tyrosine kinase inhibitors or immunotherapy are not financially affordable. Therefore, establishment of a method to identify potential responders to cytotoxic chemotherapy is imperative for such patients.
A cardinal feature of cancer in the advanced stage is the ability of cancer cells to invade surrounding tissues. Malignant cells migrate through the vascular endothelium and enter the vascular lumen where they constitute circulating tumor cells (CTCs) (12, 13). Some of these freely moving cells may dock on the vascular bed of other organs, regaining the potential of proliferation. The occurrence of distant metastasis usually indicates a terminal stage of cancer (14, 15). For unknown reasons, therapeutic responses vary widely among patients at this stage, and range from complete responses to ineffectiveness to anticancer drugs. Therefore, establishment of a reliable assay that can predict the therapeutic response of cancer patients is invaluable to oncologists.
In this study, live CTCs isolated from blood were cultured on a layer of human mesothelial cells for a drug sensitivity test to assess its feasibility in clinical application.
Materials and Methods
Patients. Thirty-one patients with advanced HCC and one patient with liver granuloma were enrolled between June 1, 2019, and June 30, 2020. Enrolled patients had either extrahepatic metastasis or main portal vein thrombosis. A total of 16 ml of peripheral blood was collected from each participant. All enrolled participants in the study had signed informed consent. This study was approved by the Medical Ethics and Human Clinical Trial Committee of the Chang Gung Memorial Hospital (IRB No. 201900333B0).
Cell culture and IC50 estimation. Mahlavu cells were cultured in Dulbecco’s Modified Eagle Medium containing 5% fetal bovine serum (Gibco, Grand Island, NY, USA). Cytotoxic chemotherapy drugs, including 5-fluorouracil, gemcitabine, oxaliplatin, and cisplatin, were serially diluted and added into the culture medium. Cells were incubated for 72 h to assess cell viability using a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide MTT assay. IC50 assay was used to determine drug concentration that achieves 50% cell death.
CTCs culture and drug sensitivity testing. Upon peripheral blood was withdrawn from HCC patients, the first 4 ml of blood were discarded to avoid epithelial contamination and the remaining blood was used for the isolation of live CTCs. The CD45 depletion procedures were performed as previously described and repeated twice to obtain CD45-depleted cells (16, 17). CTCs were divided into two equal portions to assess their growth response in the presence or absence of the tested drugs. CTCs were co-cultured with human mesothelial cells as described in previous studies (18, 19). The anticancer drug was added into one set of cultures and incubated for 72 h. Immunohistochemistry was carried out to identify CTCs. CTC slides, from patients with high AFP (>500 ng/ml), were incubated with anti-alpha-fetoprotein (AFP) antibody (Abcam Inc., Cambridge, MA, USA). Otherwise, CTC slides from patients with low AFP were incubated with an anti-glypican 3 antibody (Abcam).
Cell counting of CTC. A total of 8 ml of blood was used for CTC counting. Flow cytometry was performed to identify and quantify CTCs (EpCAM+/Hochest+/CD45–). The immunomagnetically enriched samples were labeled using an Alexa Fluor® 488-conjugated monoclonal antibody to EpCAM (1:400; Cell Signaling Technology Inc., Danvers, MA, USA) and stained using a DNA-detecting far-red-fluorescent dye (Draq5, 50 μM; Abcam). An isotype control antibody was used as an internal control. Blood samples from healthy individuals (4 ml) spiked with or without 1,000 cancer cells (purchased from Taiwan’s Food Industry Research and Development Institute, Hsinchu, Taiwan, ROC) were also included to ensure analytical accuracy. Performance recovery was defined as the proportion of detected cancer cells by flow cytometry (BD FACSCalibur; BD Biosciences, San Jose, CA, USA) to the number of spiked cells; besides a stable coefficient of variation (CV) value was calculated in a previous report (16).
Statistical analysis. The paired t-test was applied for CTC counts with or without the drug in the ex vivo drug sensitivity testing. A two-tailed p-value of less than 0.05 was considered significant. Spearman’s rank correlation was used to evaluate the association between CTC drug sensitivity and clinical response (overall survival and progression-free survival). The correlation between the longitudinal follow-up of CTC counts and clinical response was analyzed by Spearman’s correlation. A p-value <0.05 was considered highly correlated. Kaplan–Meier plotting was performed for survival analysis and the log-rank test for statistical analysis. All statistical analyses were performed by using the SPSS version 17 (SPSS Inc., Chicago, IL, USA).
Results
Clinical characteristics of patients. Table I lists the clinical information of patients receiving cytotoxic chemotherapy or targeted drug therapy. The individual numbered 1 is a patient with hepatic granuloma of unknown etiology (as a negative control). Individuals numbered 2 to 32 are patients with advanced HCC. The age range of enrolled patients was between 39 to 76 years old. Among the 31 HCC patients, 22 were HBV surface antigen (HBsAg) positive, 4 were anti-HCV antibody positive, and 6 were non-B and non-C. Of these patients, 23 had distant metastasis, which occurred in the lungs (14 cases), lymph nodes (4 cases), bone (2 cases), hepatic vein (HV, 2 cases), inferior vena cava (IVC, 2 cases), and regional nodal (1 case). Two patients had two metastatic sites. Of the 31 HCC patients, 25 had alpha-fetoprotein (AFP) >20 ng/ml, 27 had cirrhosis, 17 had ascites, 16 had more than three tumors, and 10 had main portal vein thrombosis. Hemogram and biochemistry data including leukocyte count, hemoglobin levels, platelet count, neutrophil count, alanine transaminase (ALT) activity, bilirubin levels, prothrombin levels, albumin levels, and creatinine levels are listed in Table I.
Clinical data of the 32 patients included in this study.
Culture of live CTCs and immunohistochemical staining. Live CTCs were isolated according to our previous studies (16-19). The CTCs isolated from each patient were divided into two equal parts and treated with or without an anticancer drug using ex vivo culture. The anticancer drug concentrations used in this study were based on the IC50s of the drugs tested in Mahlavu cells. The immunohistochemical staining was used to detect and identify the CTCs. As shown in Figure 1, CTCs could either grow as isolated cells (Figure 1A and B) or cell clusters (Figure 1C and D).
Immunohistochemical staining of circulating tumor cells (CTCs) after co-culture with human mesothelial cells. Cells were fixed with acetone before staining. (A) and (B) CTCs were stained using the anti-glypican 3 antibody. (C) and (D) CTCs were stained using the anti-AFP antibody. Arrow, clustered CTCs.
Cytotoxic chemotherapy drug sensitivity tests on live CTCs. Twenty advanced HCC patients (patients 2 to 21) were either under scheduled courses of chemotherapy or planning to receive the first course of chemotherapy. Table II displays detailed outcome of the CTC drug test, including the number of CTCs isolated for each patient, the anticancer drug used, the number of CTCs before and after drug treatment, and therapeutic information for patients (the chemotherapeutic drug used after CTC drug test, drug response, and time to death). Of the 20 patients, 13 (including patient 1) had never received cytotoxic chemotherapy; 8 cases had received chemotherapy before enrolling in the study. In addition, 18 patients had culturable CTCs, whereas the rest of the patients (patients 1, 20, and 21) had no CTCs to culture before drug testing. After determining which chemotherapy drug combination would be given to patients in the next course, one of the drugs was selected for sensitivity testing ex vivo. Unexpectedly, we observed that CTCs from 10/12 (83.3%) patients who had never received chemotherapy were sensitive to chemotherapeutic agents used for testing. In contrast, CTCs isolated from patients (6 of 6) who had received chemotherapy, even just one course, showed resistance to chemotherapeutic agents (Fisher’s Exact p=0.002). The Wilcoxon signed-rank test revealed a significant reduction of the numbers of CTCs after the addition of chemotherapy drugs in ex vivo cultures of CTCs isolated from patients (n=12) who had not received therapy (paired t-test, p=0.025). Conversely, the numbers of CTCs isolated from patients (n=6) who had received chemotherapy showed a significant increase after drug treatment ex vivo (paired t-test, p=0.028). Notably, patient-19 and 20, who did not have culturable CTCs, experienced a complete or partial response to cytotoxic chemotherapy.
CTC counts, culturable CTC numbers, and drug information for 21 patients.
Sorafenib sensitivity tests on live CTCs. The CTCs isolated from 11 advanced HCC patients (patients 22 to 32) were cultured ex vivo for the sorafenib sensitivity test. Table III displays the numbers of CTC (before and after the sorafenib test) and the patients’ clinical information (targeted drugs used after the sorafenib test, drug response, and time to death). Out of the 11 patients, only one (28 patients) never received HAIC or TT, while the other 10 received TT before CTC drug tests. In the absence of sorafenib, 2 of 11 patients (patients 24 and 27) had no culturable CTCs, although two CTCs were found on the slide from patient-24. The CTC drug test showed that the CTCs of 7 (7/11, 63.6%) patients were sensitive to sorafenib, whereas those from 3 (3/11, 27.3%) patients were not. The results showed that previous targeted therapy did not affect CTC counts in the ex vivo culture (Fisher’s Exact p=1.000). Except for two patients who were lost in follow-up, two patients (patients 28 and 32) expired during the study period.
CTC drug sensitivity in 11 patients.
Comparison of resistance in CTCs between patients receiving cytotoxic chemotherapy and TT. After excluding two patients (patients 20 and 21), who had a complete response and partial response to chemotherapy as well as no culturable CTCs available for assessment, we compared resistance of CTCs from patients treated with cytotoxic chemotherapy and sorafenib. For patients who had received drugs (chemotherapy, n=6; sorafenib therapy, n=8) before CTC tests, the Fisher’s exact test showed that CTCs isolated from patients receiving TT had a lower probability for drug resistance (p=0.031).
Correlation between total CTC counts and overall survival. Several previous studies have suggested that CTC counts may correlate with the survival of patients with cancers including breast, colorectal, and prostate cancer (20-22). Therefore, the correlation between CTC counts from the HCC patients who received cytotoxic chemotherapy (Table II) and overall survival was assessed. Spearman’s correlation analysis showed that CTC counts were negatively associated with the overall survival of patients (p=0.001). Additionally, Kaplan– Meier analysis exhibited a significant poor survival rate in patients with higher CTC count (n=10, >30 cells/ml) than lower CTC count (n=10, <30 cells/ml) (p=0.016) (Figure 2).
Kaplan–Meier analysis for overall survival in relation to circulating tumor cell (CTC) counts of. Blue line, overall survival of patients with CTCs <30 cells/ml; Green line, overall survival of patients with CTCs >30 cells/ml.
Discussion
Despite recent advances in anticancer drugs, HCC patients with distant metastasis do not benefit significantly from these treatments (23-25). One of the main reasons is that most patients with advanced HCC do not respond well to anticancer drugs. Besides, no effective biomarkers are available to identify a small subset of patients who might respond and there is no satisfactory way to determine which drugs to use. In this study, we developed an ex vivo CTC drug sensitivity test to predict therapy responses in patients with advanced HCC. To our surprise, we discovered that 83.3% of CTCs isolated from patients who had never received chemotherapy were sensitive to chemotherapeutic drugs regardless of which ones. However, CTCs isolated from all patients who had received cytotoxic chemotherapy (with culturable CTCs) developed resistance to chemotherapeutic drugs in the ex vivo culture. The results implied that once advanced HCC patients receive any type of cytotoxic chemotherapy, their cancer cells, especially CTCs, rapidly develop resistance. According to our present data, chemotherapy might only be effective in the treatment of naïve HCC patients. Furthermore, a type of patients (such as patient-19 and 20) who had received chemotherapy drugs without culturable CTCs had a complete or partial response, implying that such patients may benefit from chemotherapy. At present, the reason for the ex vivo drug-dependence-like phenotype is unclear.
Unlike cytotoxic chemotherapeutics, the sorafenib sensitivity test did not show a significant effect on the development of resistance of CTCs isolated from patients who had received targeted drugs. Clinically, targeted therapy has been demonstrated to prolong overall survival of patients with advanced HCC (26). Nevertheless, cytotoxic chemotherapy drugs are more available and affordable than targeted therapy drugs in developing countries (27). Therefore, chemotherapy remains the first choice for advanced HCC patients with financial problems. In this study, although the patients who received targeted therapy had no objective tumor response (complete or partial), 8 of 9 (88.9%) patients demonstrated a longer survival time (>200 days), compared to patients (1 of 17, 5.9%) who received chemotherapy. The rapid development of drug resistance may be the reason why patients with advanced HCC who received cytotoxic chemotherapy had shorter overall survival than those who received targeted therapy.
CTC quantification has been reported to have the potential to predict survival in cancer patients (28). For example, a previous study indicated that the CTC count could serve as a predictive biomarker for overall survival in patients with castration-resistant prostate cancer who received docetaxel chemotherapy (29). Also, a clinical trial indicated that CTC-positive patients with triple-negative breast cancer have a poorer prognosis in disease-free survival and overall survival (30). In this study, the CTC count had a similar value in predicting the overall survival of patients with advanced HCC who received chemotherapy. Regardless of the drug and number of courses given, CTC counts were negatively associated with overall survival of patients. This finding argued that cytotoxic chemotherapy did not significantly affect the outcomes in these advanced HCC patients. If the ex vivo culture results can reflect what happens in the human body, it can be deduced that advanced HCC patients develop resistance to cytotoxic therapeutics much faster than we thought. In case CTCs are still present following initial cytotoxic chemotherapy, they are very likely already drug resistant. On the other hand, is CTCs are less likely to develop resistance to sorafenib. Therefore, we could adjust the anticancer strategy for advanced HCC patients accordingly. It may be necessary to measure the number of CTCs in the blood before and after initial chemotherapy. If CTCs persist, different anticancer drugs such as targeted drugs should be tried.
In conclusion, in this study we developed an ex vivo culture system to test drug sensitivity for CTCs isolated from advanced HCC patients. We found that CTCs isolated from patients receiving cytotoxic chemotherapy rapidly developed resistance, whereas CTCs from patients receiving sorafenib barely developed resistance. The results suggested that for patients with advanced HCC who choose chemotherapy, prior drug sensitivity test of CTCs may be beneficial for predicting treatment response.
Acknowledgements
This work was supported by the Chang Gung Memorial Hospital, Linkou, Taoyuan, Taiwan, ROC under Grants CMRPG3J0681, CMRPG3J0682, and CMRPG3J0683. The authors are very grateful for the technical assistance from members of Liver Research Center, Chang Gung Memorial Hospital, Linkou, Taoyuan, Taiwan, ROC.
Footnotes
Authors’ Contributions
CTY designed and supervised the study; CHH, CTY, YHH, and MWL drafted the manuscript; CHH, CTY, YHH, and MWL collected clinical data and interpreted the data; CTY performed statistical analysis.
Conflicts of Interest
All Authors declare no conflicts of interest regarding this study.
- Received March 7, 2022.
- Revision received March 30, 2022.
- Accepted March 31, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.