Figures
Abstract
Objective
Hepatocellular carcinoma (HCC) and its recurrence are major problems in living donor liver transplantation (LDLT). Several biomarkers have been used to investigate this event. We conducted a prospective controlled study to determine the activities of the basic fibroblast growth factor (FGF-2), survivin, Ki67, endostatin, and vascular endothelial growth factor (VEGF) in different conditions before, early after, and late after LDLT with and without HCC recurrence.
Methods
Fifty patients with virus-related HCC who underwent LDLT were enrolled in this 2-year cross-sectional study. During the study period, recurrent HCC was identified in 9 patients (study group, n = 9) and 41 patients (control group, n = 41) had no recurrence after LDLT. The mean time to HCC recurrence was 587.11 ± 398.64 days (range, 90–1352 days). Microvascular invasion (MVI) was found in 66.7% (6/9) of the recipients, as determined on pathological examination. The serum biomarkers were investigated by using enzyme-linked immunosorbent assay at the different LDLT stages.
Results
The serum levels of the biomarkers significantly correlated with LDLT and HCC recurrence in the repeated-measures analysis (F = 31.676, P = 0.000). Significant differences were observed in the effects of all biomarkers (F = 85.313, P = 0.000) and the time to HCC recurrence after LDLT (F = 3.178, P = 0.046). The biomarkers, ordered by the observed power of the test for HCC recurrence after LDLT, were FGF-2 (1.000) > survivin (0.999) > Ki67 (0.949) > endostatin (0.411) > VEGF (0.305).
Conclusions
Different biomarker activities may be implicated in the pathogenesis of HCC recurrence after LDLT. Oncogenes may not exist in the new graft but may still be present in the peripheral blood. The timing of HCC recurrence and impact of MVI in the explanted liver requires confirmation in larger studies with a longer follow-up.
Citation: Chiu K-W, Nakano T, Chen K-D, Hsu L-W, Lai C-Y, Huang C-Y, et al. (2015) Repeated-Measures Implication of Hepatocellular Carcinoma Biomarkers in Living Donor Liver Transplantation. PLoS ONE 10(5): e0124943. https://doi.org/10.1371/journal.pone.0124943
Academic Editor: Kwan Man, The University of Hong Kong, HONG KONG
Received: February 1, 2015; Accepted: March 20, 2015; Published: May 15, 2015
Copyright: © 2015 Chiu et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Data Availability: All relevant data are within the paper.
Funding: This work was supported by grant number CMRPG8B0921 from the Chang Gung Memorial Hospital of Taiwan.
Competing interests: The authors have declared that no competing interests exist.
Introduction
In Taiwan, hepatocellular carcinoma (HCC) is a leading cause of cancer mortality in male patients. Living donor liver transplantation (LDLT) was reported to be potentially curative for cirrhotic patients with HCC, as it simultaneously treats both conditions [1–2]. In the liver transplantations performed at our center, abnormal liver function, acute rejection, recurrent infectious disease, or transient portal hypertension esophageal bleeding were complications observed after LDLT [3–5]. However, one of the most serious complications of LDLT is recurrent HCC, especially when Edmondson-Steiner’s grading indicates poor differentiation [6–7]. This is an issue not only for our programs but also for liver transplantation centers worldwide. In addition to alpha-fetoprotein (AFP), a number of specific biomarkers provide useful clinical information for early detection and prognosis, pathogenesis, and treatment efficacy [8]. Many tumor markers have been used as prognostic and therapeutic biomarkers. The basic fibroblast growth factor (FGF-2) is a member of the fibroblast growth factor family. It has been hypothesized that during both wound healing of normal tissues and tumor development, the action of heparan sulfate-degrading enzymes activates FGF-2, thus mediating the formation of new blood vessels in a process known as angiogenesis. Survivin is a unique member of the inhibitor of apoptosis protein family and regulates the cell cycle in most tumors. However, it is barely detectable in terminally differentiated normal cells and tissues. Differential expression of survivin in cancer compared with normal tissues makes it a useful tool in cancer diagnosis and a promising therapeutic target. The Ki-67 protein is a cellular marker for proliferation and can be used to identify the actively dividing fraction of a given cell population. The fraction of Ki-67-positive tumor cells is often correlated with the clinical course of cancer. Endostatin is a broad-spectrum angiogenesis inhibitor and may interfere with the pro-angiogenic action of growth factors such as FGF-2 and vascular endothelial growth factor (VEGF). VEGF is the most important angiogenesis factor and a highly specific mitogen for blood vessel endothelial cells. It stimulates microvessel endothelial cells to proliferate and increase permeability, resulting in tumor angiogenesis [9–13]. Based on the pathogenesis of HCC, we conducted this prospective, cross-sectional, controlled study by using FGF-2, survivin, Ki67, endostatin, and VEGF activities to determine HCC recurrence before, early after, and late after LDLT.
Materials and Methods
Patients
This 2-year prospective study enrolled 50 HCC-associated patients who underwent LDLT in our liver transplant programs. Before LDLT, all non-HCC-associated patients, including pediatric cases, were excluded from this study. The study group comprised nine recipients with HCC recurrence within the 2-year period after LDLT. The control group consisted of 41 recipients without HCC recurrence after LDLT. The mean age of the two groups was 58.0 and 59.0 years, respectively. The male-to-female ratios in the two groups were 9:0 and 38:3, respectively. The mean time to HCC recurrence was 587.11 ± 398.64 days (range, 90–1352 days). Microvascular invasion (MVI) to the pathological site was found in 66.7% recipients (6/9). Edmondson-Steiner’s cell grading of HCC showed that 88.9% (8/9) of patients had grade II cell differentiation. HCC recurred in 88.9% (8/9) of the cases with distant metastases, including the lung (55.6%, 5/9), adrenal gland (22.2%, 2/9), lymph nodes (22.2%, 2/9), brain (11.1%, 1/9), and bone (11.1%, 1/9). In HCC recurrence after LDLT, AFP levels remained within normal limits for 44.4% (4/9) of cases, was 2–4 times above the normal limit in 22.2% (2/9), and was more than several thousand nanograms per milliliter in 33.3% (3/9) (Table 1). At follow-up for all the study cases, the survival rate was 100% (41/41) in the control group. In contrast, 44.4% (4/9) of the patients in the study group did not survive, even with advanced treatment, and 55.6% (5/9) survived until the end of the study. Patient serum samples were analyzed by using enzyme-linked immunosorbent assay (ELISA), focusing on FGF-2, survivin, Ki67, endostatin, and VEGF at three different time periods as follows: before LDLT (preoperative day 1 as the baseline before transplantation), early after LDLT (postoperative day 1 as the baseline after liver transplantation), and late after LDLT (control group, postoperative day 30; study group, the day of the first detection of HCC recurrence during follow-up). For cases of HCC recurrence after LDLT, blood samples for before and early after LDLT were obtained from a tissue bank of our liver transplant programs within the study period and analyzed. Samples for late after LDLT were obtained at the time of documentation of HCC recurrence after LDLT. HCC patients with high AFP levels (>200 ng/mL) and poor differentiation (>grade III in Edmondson-Steiner’s cell grading) were not considered for LDLT in our program [2]. As a result, AFP was not used in our biomarker investigation. Our follow-up protocol for detection of HCC and distant metastases consisted of regular blood tests, including liver function and AFP analysis every month, ultrasonography and chest radiography every 3 months, and computed tomography of the liver and chest every year. In additional to the chest and brain magnetic resonance imaging, computed tomography was performed when the AFP level was elevated or ultrasonography and/or chest radiography results showed abnormal findings.
Fibroblast growth factor 2
Serum FGF-2 levels were measured by using an ELISA measurement kit (R&D Systems, Minneapolis, MN, USA). Each well containing a precipitate was mixed well with 100 μL of the assay diluent HD1-46. Next, 150 μL of standard, control, or sample specimen was added to each well. Wells were covered with the provided adhesive strip and incubated for 3 h at room temperature (18–23°C). A plate layout was provided as a record of the standards, and the samples were assayed. FGF basic conjugate (200 μL) was added to each well, the plate was covered with a new adhesive strip, and incubated for 2 h at room temperature. Substrate (50 μL) was added to each well, covered with a new adhesive strip, and incubated for 45 min at room temperature. Amplifier solution (50 μL) was added to each well, and the plate was covered with a new adhesive strip. Finally, 50 μL of stop solution was added to each well. The optical density of each well was determined within 30 min. Absorbance at 450 nm of the colored reaction product was analyzed with an ELISA microplate reader. All serum samples were assayed in duplicate by one operator in order to assess interassay precision. Results were calculated from a standard curve generated by a parametric logistic curve fit and expressed in picogram per milliliter serum.
Survivin
Serum survivin was measured manually by using ELISA with Quantikine kits (R&D Systems). Standard dilutent (100 μL) was pipetted into the appropriate wells, and the plate was gently tapped to mix the contents. The plate was sealed and incubated at room temperature on a plate shaker for 1 h at 500 rpm. Next, 100 μL of antibody was pipetted into each well, and the plate was sealed and incubated at room temperature on a plate shaker for 1 h at 500 rpm. Conjugate (100 μL) was added to each well, and the plate was sealed and incubated at room temperature on a plate shaker for 30 min at 500 rpm. Substrate (100 μL) was pipetted into each well and incubated for 30 min at room temperature on a plate shaker at 500 rpm. Finally, 100 μL of stop solution 2 was pipetted into each well. The optical density at 450 nm was analyzed by using a LabSystems Multiskan Spectrum Microplate Reader (Thermo Scientific, Waltham, MA, USA), according to the manufacturer’s protocol. Results were calculated from a standard curve generated by a parametric logistic curve fit and expressed in pictogram per milliliter serum.
Ki67
Serum Ki67 was measured by using a commercially available Human Antigen Ki-67 ELISA kit (Cusabio Biotech Co., LTD, Wuhan, Hubei, China) based on a sandwich ELISA. Results were calculated from a standard curve generated by a parametric logistic curve fit and expressed in nanograms per milliliter serum. Standard (100 μL) and sample specimens were added to each well and incubated for 2 h at 37°C. Next, 100 μL of prepared detection reagent A was added and incubated for 1 h at 37°C. Detection reagent B (100 μL) was added and incubated for 30 min at 37°C. Substrate (90 μL) was added and incubated for 15–25 min at 37°C. Finally, 50 μL of stop solution was added. Absorbance at 450 nm was immediately analyzed in a microplate reader.
Endostatin
Serum endostatin was measured by using a commercially available ELISA kit for endostatin (R&D Systems). Results were calculated from a standard curve generated by a parametric logistic curve fit and expressed in nanograms per milliliter serum. Samples were mixed with 100 μL of assay diluent RD1W, covered with the provided adhesive strip and incubated for 2 h at room temperature on a horizontal orbital microplate shaker set at 500 rpm. The standards were recorded on a plate layout, and the samples were assayed. Next, conjugated endostatin (200 μL) was added to each well, and the plate was covered with a new adhesive strip and incubated for 2 h at room temperature on a shaker. Substrate (200 μL) was added to each well and incubated in the dark for 30 min at room temperature. The reaction was stopped with 50 μL of stop solution, and the plate was tapped to ensure thorough mixing. The optical density at 450 nm of each well was determined within 30 min by using a microplate reader.
Vascular endothelial growth factor
VEGF level was measured by using a commercial ELISA kit for VEGF (R&D Systems), following the manufacturer’s instructions. Results from a standard curve generated by a parametric curve fit were calculated and expressed in pictogram per milliliter serum. An ELISA plate was coated with 100-μL/well capture antibody overnight at room temperature. Each well was washed with wash buffer (400 μL) by using a squirt bottle and an electrical single channel pipette. Plates were blocked by adding to each well 300 μL of phosphate-buffered saline containing 1% bovine serum albumin, 5% sucrose, and 0.05% NaN3 and incubated at room temperature for at least 1 h. The samples and standards were diluted in polypropylene tubes, and 100 μL of sample or standard was added to each well. The plate was mixed by gentle tapping for 1 min, covered with an adhesive strip, and incubated for 2 h at room temperature. A biotinylated detection antibody was diluted, and 100 μL was added to each well. The plate was covered with a new adhesive strip and incubated for 2 h at room temperature. Streptavidin conjugated with horseradish peroxidase (100 μL) was added to each well, and the plate was covered and incubated for 20 min at room temperature. Substrate (100 μL) was added to each well and incubated in the dark for 20–30 min at room temperature. Finally, 50 μL of stop solution was added to each well, and the plate was gently tapped to ensure thorough mixing. The optical density of each well at 450 nm was determined within 30 min by using a microplate reader.
For each assay, baseline and post-intervention samples from the same individual were analyzed in the same experiment. Each sample was assayed in duplicate by one operator to assess interassay precision.
Ethics statement
All of the participants provided written informed consent to participate in this study. The ethics committees approved all consent procedures. Ethical approval was obtained for association and institutional standards by the institute review board of Chang Gung Memorial Hospital. The ethics committee specifically approved the complete study (No. 103-0352C). None of the transplant donors were from a vulnerable population, and all of the donors or their next of kin freely provided written informed consent.
Statistical analysis
Data were analyzed by using the Statistical Package for repeated measures (SPSS v. 22.0 for Windows, IBM Corp., Armonk, NY, USA). Results were obtained from comparisons of between- and within-subject effects, partial η2, and observed power. The Mann-Whitney U test was used to compare the difference in biomarkers between the patients with and those without HCC recurrence. Differences were considered statistically significant when P < 0.05.
Results
The levels of the HCC biomarkers FGF-2, survivin, Ki67, endostatin, and VEGF between the study and control groups before, early after, and late after LDLT were analyzed by ELISA. The Mann-Whitney U test results showed a significant difference between the study (recurrent HCC) and control (no recurrence) groups before, early after, and late after LDLT for FGF-2 (-4.14 ± 2.38 vs. -11.25 ± 4.98 pg/mL, P = 0.000; -4.45 ± 3.68 vs. -10.20 ± 4.19 pg/mL, P = 0.000; 0.91 ± 3.85 vs. -8.22 ± 4.76, P = 0.000, respectively), survivin (-1261.09 ± 68.89 vs. -266.44 ± 519.86 pg/mL, P = 0.000; -1275.07 ± 59.25 vs. -291.39 ± 476.80 pg/mL, P = 0.000; -1250.34 ± 70.57 vs. -209.19 ± 593.00 pg/mL, P = 0.000, respectively), and Ki67 (0.02 ± 0.68 vs. -1.45 ± 1.02 ng/mL, P = 0.001; -0.56 ± 0.40 vs. -1.90 ± 0.76 ng/mL, P = 0.000; -0.30 ± 0.56 vs. -1.41 ± 0.87 ng/mL, P = 0.003, respectively). Differences in VEGF before and early after LDLT were also significant (119.29 ± 53.10 vs. 278.37 ± 195.49 ng/mL, P = 0.033; 128.92 ± 71.85 vs. 406.45 ± 338.43 ng/mL, P = 0.014). However, no significant differences were observed between the patients with and those without HCC recurrence in terms of endostatin level before, early after, or late after LDLT (2.23 ± 2.68 vs. 3.07 ± 1.45 ng/mL, P = 0.210; 2.23 ± 2.35 vs. 3.06 ± 1.47 ng/mL, P = 0.108; 6.69 ± 4.76 vs. 4.50 ± 3.01 ng/mL, P = 0.293, respectively) or in VEGF late after LDLT (229.70 ± 85.28 vs. 418.52 ± 378.24 pg/mL, P = 0.503) (Figs 1–5).
The Mann-Whitney U test was used to compare the differences in the biomarkers at the various time points between the recipients with and those without HCC recurrence.
The Mann-Whitney U test was used to compare the differences in the biomarkers at the various time points between the recipients with and those without HCC recurrence.
The Mann-Whitney U test was used to compare the differences in the biomarkers at the various time points between the recipients with and those without HCC recurrence.
The Mann-Whitney U test was used to compare the differences in the biomarkers at the various time points between the recipients with and those without HCC recurrence.
The Mann-Whitney U test was used to compare the differences in the biomarkers at the various time points between the recipients with and those without HCC recurrence.
The observed power of the repeated-measures analysis of the biomarker levels late after LDLT with HCC recurrence showed that the power of FGF-2 (1.000) was > survivin (0.999) > Ki67 (0.949) > endostatin (0.411) > VEGF (0.305) (Table 2). Tests of the between-subjects effects on HCC recurrence revealed that all biomarkers were significant (F = 31.676, P = 0.000, partial η2 = 0.398, observed power = 1.000) (Table 3). The results of the tests of within-subjects effects showed a strong correlation to the biomarkers (F = 85.313, P = 0.000, partial η2 = 0.640, observed power = 1.000), but a weak correlation to time (F = 3.178, P = 0.046, partial η2 = 0.062, observed power = 0.596) (Table 4).
Discussion
Based on the pathogenesis of HCC, the identification of significant factors and the mechanisms of HCC recurrence after LDLT remain major challenges in contemporary biomedicine. In the period between LDLT and HCC recurrence, timing after transplantation was found to be a potential influencing factor and is one of the factors investigated here. In our study, the mean time to HCC recurrence was 587.11 ± 398.64 days. Liver grafts were obtained from donors without HCC, and the circulating individual oncogene could be identified from peripheral blood but not from the liver graft at the time of HCC recurrence. We used five biomarkers reported to be associated to HCC to investigate HCC recurrence at different times after LDLT. These biomarkers were the growth factor FGF-2, the angiogenesis factors VEGF and endostatin, and the apoptosis- and cancer cell proliferation-related proteins survivin and Ki67. A significant difference was observed between the groups with and without HCC recurrence in terms of serum FGF-2, survivin, and Ki67 levels at different time points (P < 0.000). This is further evidence that HCC recurrence at different time points is associated with the activity of different biomarkers in circulating blood [14]. In our recent study, in donor grafts, specific protein expression did not change in the peripheral blood in the LDLT recipients [15] and was homogenous in the new liver [16–18]. One possible reason is that most HCC recurrences in the present study were isolated distant metastases (77.8%). Although VEGF has been documented in HCC development and prognosis [13], VEGF activity did not obscure differences in the comparison between the study and control groups in an individual paired t test analysis. For both HCC recurrence and liver graft regeneration after LDLT, angiogenesis and the VEGF receptor pathways might be activated, possibly resulting in the loosening of intercellular junctions of endothelial cells and promoting angiogenesis. The VEGF receptor pathway plays a physiological role that may not differentiate between HCC recurrence and liver graft regeneration after LDLT [19–20]. In our experience, liver graft regeneration requires significant blood flow [21] and leads to the activation of this pathway. Studies performed under the same conditions suggest that serum endostatin is also significantly related to liver regeneration in an animal model [22]. To determine the effect of biomarkers on HCC recurrence over time, a repeated-measures analysis of variance has been recommended as the best method of investigation [23]. Analysis of variance is a generalized method of the paired t test. The only difference between these methods is that, for the former, measurements were taken in the same individuals. The measurements are likely to be correlated, and any analysis must consider such correlations. The advantages of this statistical method are that subjects are also the controls themselves, isolating variability between subjects, and the analysis can focus more precisely on recurrent effects. As the present study is a 2-year prospective study with a limited number of cases of HCC recurrence after LDLT, the repeated-measures design is suitable. Because the subjects were also the controls themselves, fewer subjects are needed [24]. A repeated-measures analysis of variance (ANOVA) may be used when the same parameter has been measured under different conditions in the same subjects. Subjects can be divided into different groups, as in the present study. Performing a repeated-measures analysis can overcome the statistical limitations in cases with HCC recurrence over time and reduce bias from individual biomarker analyses. In the tests of between-subjects effects, all of the biomarkers reached a level of significance (0.000) and were involved in HCC recurrences after LDLT, with an observed power of up to 1.000 (Table 3). We demonstrated how invariance between and within subjects can be tested by high-frequency repeated measures in this study. Because individual biomarker activities may be expressed differently after LDLT in individual recipients over time, a simple paired t test would be a biased description of this phenomenon. Meanwhile, the results of the tests of within-subjects effects in our study showed that the biomarkers were significantly involved in HCC recurrence after LDLT and further showed that time was one of the influencing factors (Table 4). This approach also solved the bias problem in studies with a small number of cases. When HCC recurs late after LDLT, this can be determined from the observed biomarker levels that reflect the bioactivity of the liver [21–22]. Measures of effect size in repeated-measures ANOVA indicate the degree of association between an effect and its dependent variable. Partial η2 is a common measure of effect size and estimates of the degree of association of the sample. In Table 2, partial η2 was 0.375 in FGF-2_3, representing an effect size of 37.5% in the relationship between the value and HCC recurrence. The observed power was 1.000, much higher than that of other biomarkers. In this study, the bioactivity of FGF-2 was > survivin > Ki67 > endostatin > VEGF during HCC recurrence after LDLT (Table 2). Although we found that MVI does not influence patient outcome after LDLT [1], 77.8% of patients with HCC recurrence presented with distant metastasis, of which 66.7% were associated with MVI in the explanted liver (Table 1). Given that circulation biomarkers have been previously identified from peripheral blood, it is likely that MVI influences recurrence after LDLT [25–26]. A recent study showed that higher VEGF plasma levels before liver transplantation may influence HCC recurrence [27]. In our study, we found a significant difference (P = 0.033) in serum VEGF level by using the Mann-Whitney U test of the data of the samples obtained before LDLT. After liver transplantation, immunosuppressive agents such as sirolimus downregulate mRNA expression of VEGF, but not that of FGF-2 [28]. This may be one of the reasons for the lack of significant differences in VEGF between samples collected early and late after LDLT. In contrast, FGF-2 level was found to have a higher observed power in the tests of between-subjects effects in all the studied biomarkers. Survivin is critical in promoting cell proliferation [29], and all cell lines stain strongly for Ki67 [30]. These three biomarkers could be important biomarkers for detection of HCC recurrence after LDLT.
In conclusion, the present study indicates that HCC recurrence is defined by the activity of biomarkers such as FGF-2, survivin, Ki67, endostatin, and VEGF. This activity is naturally differs between individuals and becomes more complex over time. It is difficult to predict HCC recurrence with a single variable. However, a repeated-measures analysis shows the fluctuation of multiple biomarkers over time after LDLR and may predict HCC recurrence and pathogenesis. Oncogenes are not present in the new graft but appear to remain in peripheral blood. The timing of HCC recurrence and the impact of MVI in the explanted liver requires confirmation in larger studies with longer follow-up periods.
Acknowledgments
This work was appreciatively assisted by Ms. Mei-Chin Hsu for the statistical analysis. This work was supported by grant number CMRPG8B0921 from the Chang Gung Memorial Hospital of Taiwan.
Author Contributions
Conceived and designed the experiments: KWC CLC. Performed the experiments: CYL CYH. Analyzed the data: KWC TN KDC SG. Contributed reagents/materials/analysis tools: KWC TN LWH YFC. Wrote the paper: KWC TN SG.
References
- 1. Chen CL, Concejero AM. Liver transplantation for hepatocellular carcinoma in the world: the Taiwan experience. J Hepatobiliary Pancreat Sci. 2010; 17: 555–558. pmid:19760360
- 2. Chen CL, Kabiling CS, Concejero AM. Why does living donor liver transplantation flourish in Asia? Nat Rev Gastroenterol Hepatol. 2013; 10: 746–751. pmid:24100300
- 3. Chiu KW, Chen YS, de Villa VH, Wang CC, Eng HL, Wang SH, et al. Characterization of liver enzymes on living related liver transplantation patients with acute rejection. Hepatogastroenterology. 2005; 52: 1825–1827. pmid:16334785.
- 4. Lu LS, Hung FC, Chuah SK, Chiu KW, Hu TH, Cheng YF, et al. Recurrent gastric varices bleeding after living-related liver transplantation successfully eradicated by splenic artery embolization. Hepatogastroenterology. 2012; 59: 1586–1588. pmid:22414565
- 5. Chiu KW, Nakano T, Hu TH, Tseng HP, Cheng YF, Jawan B, et al. Influence of CYP2C19 genotypes on graft pathological findings and postoperative liver function in recipients after living-donor liver transplantation. Ann Transplant. 2010; 15: 38–43. pmid:21183874
- 6. Chiu KW, Chang-Chien CS, Chen L, Liaw YF. Ultrasonically-guided needle aspiration with preparation of cell blocks in the diagnosis of liver tumors. Hepatogastroenterology. 1994; 41: 30–33. pmid:8175110
- 7. Chiu KW, Changchien CS, Chen L. Cellular interpretation of hepatocellular carcinoma by using cell blocks technique. Hepatogastroenterology. 2002; 49: 1386–1389. pmid:12239948
- 8. El-Attar HA, Kandil MH, El-Kerm YM, El-Ghandour MK. Comparison of serum survivin and alpha fetoprotein in Egyptian patients with hepatocellular carcinoma associated with hepatitis C viral infection. Asian Pac J Cancer Prev. 2010; 11: 897–903. pmid:21133598
- 9. Schmilovitz-Weiss H, Tobar A, Halpern M, Levy I, Shabtai E, Ben-Ari Z. Tissue expression of squamous cellular carcinoma antigen andKi67 in hepatocellular carcinoma-correlation with prognosis: a historical prospective study. Diagn Pathol. 2011; 6: 121. pmid:22151825
- 10. Sharma BK, Srinivasan R, Kapil S, Singla B, Saini N, Chawla YK, et al. Serum levels of angiogenic and anti-angiogenic factors: their prognostic relevance in locally advanced hepatocellular carcinoma. Mol Cell Biochem. 2013; 383 (1–2): 103–112. pmid:24013786
- 11. Sandhu DS, Baichoo E, Roberts LR. Fibroblast growth factor signaling in liver carcinogenesis. Hepatology. 2014; 59: 1166–1173. Review. pmid:24716202
- 12. Harding JJ, Abou-Alfa GK. Predicting responsiveness to sorafenib: can the determination of FGF3/FGF4 amplifications enrich for clinical benefit? Hepatobiliary Surg Nutr. 2014; 3: 168–171. pmid:25202691
- 13. Suh YG, Lee EJ, Cha H, Yang SH, Seong J. Prognostic values of vascular endothelial growth factor and matrix metalloproteinase-2 in hepatocellular carcinoma after radiotherapy. Dig Dis. 2014; 32: 725–732. pmid:25376290
- 14. Koch I, Hofschneider PH, Lottspeich F, Eckerskorn C, Koshy R. Tumour-related expression of a translation-elongation factor-like protein. Oncogene. 1990; 5: 839–843. pmid:2193292
- 15. Chiu KW, Tai WC, Nakano T, Tseng HP, Cheng YF, Jawan B, et al. Donor graft does not affect P450 2C19 genotype expressed on peripheral blood in recipient with living donor liver transplantation. Clinical Transplant. 2010; 24: 830–834. pmid:20236133
- 16. Chiu KW, Nakano T, Hu TH, Tseng HP, Cheng YF, Jawan B, et al. Homogenous phenomenon of graft liver CYP2C19 genotypes after living donor liver transplantation. Eur J Clin Invest. 2012; 42: 352–356. pmid:21913914
- 17. Chiu KW, Nakano T, Chen KD, Hsu LW, Lai CY, Chiu HC, et al. Pyrosequencing to identify homogeneous phenomenon when using recipients/donors with different CYP3A5*3 genotypes in living donor liver transplantation. PLoS One. 2013; 8: e71314. pmid:23951129
- 18. Chiu KW, Nakano T, Chen KD, Hsu LW, Lai CY, Chiu HC, et al. Homogeneous phenomenon of the graft when using different genotype characteristic of recipients/donors in the living donor liver transplantation. World J Hepatol. 2013; 5: 642–648. pmid:24303093
- 19. Shen C, Kaelin WG Jr. The VHL/HIF axis in clear cell renal carcinoma. Semin Cancer Biol. 2013; 23: 18–25. pmid:22705278
- 20. Digklia A, Voutsadakis IA. Combinations of vascular endothelial growth factor pathway inhibitors with metronomic chemotherapy: Rational and current status. World J Exp Med. 2014; 4: 58–67. pmid:25414818
- 21. Cheng YF, Huang TL, Chen TY, Tsang LL, Ou HY, Yu CY, et al. Liver graft regeneration in right lobe adult living donor livertransplantation. Am J Transplant. 2009; 9: 1382–1388. pmid:19459827
- 22. Colakoglu T, Keskek M, Colakoglu S, Can B, Sayek I. Serum endostatin levels and regenerative capacities of normal and cirrhotic livers following partial hepatectomy in mice: the response to different resection sizes. J Surg Res. 2007; 143: 337–343. pmid:17574579
- 23. Adolf J, Schuurman NK, Borkenau P, Borsboom D, Dolan CV. Measurement invariance within and between individuals: a distinct problem in testing the equivalence of intra- and inter-individual model structures. Front Psychol. 2014; 5: 883. pmid:25346701
- 24. Shuttleworth Martyn (2009-11-26). Repeated measures design. Experiment-resources.com. Retrieved 2013-09-02.
- 25. Nishio T, Hatano E, Sakurai T, Taura K, Okuno M, Kasai Y, et al. Impact of hepatic steatosis on disease-free survival in patients with non-B non-C hepatocellular carcinoma undergoing hepatic resection. Ann Surg Oncol. Epub 2014 Nov 14.
- 26. Sohn W, Paik YH, Cho JY, Ahn JM, Choi GS, Kim JM, et al. Influence of hepatitis B virus reactivation on the recurrence of HBV-related hepatocellular carcinoma after curative resection in patients with low viral load. J Viral Hepat. Epub 2014 Nov 7.
- 27. Zhang W, Kim R, Quintini C, Hashimoto K, Fujiki M, Diago T, et al. Prognostic role of plasma vascular endothelial growth factor in patients with hepatocellular carcinoma undergoing liver transplantation. Liver Transpl. Epub 2014 Oct 6.
- 28. Wang Z, Zhou J, Fan J, Tan CJ, Qiu SJ, Yu Y, et al. Sirolimus inhibits the growth and metastatic progression of hepatocellular carcinoma. J Cancer Res Clin Oncol. 2009; 135: 715–722. pmid:19002496
- 29. Or YY, Chow AK, Ng L, Fan ST, Yau TC, Poon RT, et al. Survivin depletion inhibits tumor growth and enhances chemosensitivity inhepatocellular carcinoma. Mol Med Rep. 2014; 10: 2025–2030. pmid:25070628
- 30. Xin H, Wang K, Hu G, Xie F, Ouyang K, Tang X, et al. Establishment and characterization of 7 novel hepatocellular carcinoma cell lines from patient-derived tumor xenografts. PLoS One. 2014; 9: e85308. pmid:24416385