Protein phosphatase magnesium-dependent 1δ (PPM1D) is an oncogene, overexpressed in many solid tumors, including ovarian cancer and breast cancer. The current study examined the expression and the prognostic value of PPM1D mRNA in human hepatocellular carcinoma (HCC).
Total RNA was extracted from 86 HCC and paired non-cancerous liver tissues. PPM1D mRNA expression was determined by real-time quantitative reverse transcriptase-polymerase chain reaction (qPCR). Immunohistochemistry assay was used to verify the expression of ppm1d protein in the HCC and non-cancerous liver tissues. HCC patients were grouped according to PPM1D mRNA expression with the average PPM1D mRNA level in non-cancerous liver tissue samples as the cut-off. Correlations between clinicopathologic variables, overall survival and PPM1D mRNA expression were analyzed.
PPM1D mRNA was significantly higher in HCC than in the paired non-cancerous tissue (p<0.01). This was confirmed by ppm1d staining. 56 patients were classified as high expression group and the other 30 patients were categorized as low expression group. There were significant differences between the two groups in term of alpha-fetoprotein (α-FP) level (p<0.01), tumor size (p<0.01), TNM stage (p<0.01), recurrence incidence (p<0.01) and family history of liver cancer (p<0.01). The current study failed to find significant differences between the two groups in the following clinical characteristics: age, gender, portal vein invasion, lymphnode metastasis, hepatitis B virus (HBV) infection and alcohol intake. Survival time of high expression group was significantly shorter than that of low expression group (median survival, 13 months and 32 months, respectively, p<0.01).
Citation: Li G-B, Zhang X-L, Yuan L, Jiao Q-Q, Liu D-J, Liu J (2013) Protein Phosphatase Magnesium-Dependent 1δ (PPM1D) mRNA Expression Is a Prognosis Marker for Hepatocellular Carcinoma. PLoS ONE8(3): e60775. https://doi.org/10.1371/journal.pone.0060775
Editor: Rajeev Samant, University of Alabama at Birmingham, United States of America
Received: September 20, 2012; Accepted: March 2, 2013; Published: March 28, 2013
Copyright: © 2013 Li 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.
Funding: The authors have no support or funding to report.
Competing interests: The authors have declared that no competing interests exist.
Hepatocellular carcinoma (HCC) is the fifth most common cancer , and the third leading cause of cancer-related death worldwide. Surgical resection of the tumor offers a chance for cure, but does not achieve optimal long-term survival, mainly due to recurrence and metastasis. It is important to find the prognostic predictors for HCC. Much effort has made to identify the molecular markers correlating with the survival of HCC patients. But there is still lack of specific prognostic indicators.
In many cases, HCC develops on a background of an underlying liver disease such as chronic hepatitis, alcoholism, and exposure to hepatotoxins, . Risk factors such as hepatitis B virus (HBV) infection could alter the expression of a variety of tumor suppressor genes including p53, . The tumor suppressor gene p53 plays a major role in hepatocarcinogenesis, , . Mutation and inactivation of wild type p53 is associated with hepatocarcinogenesis. P53 mutation or inactivation preferentially occurs in moderately and poorly differentiated HCC.
Protein phosphatase magnesium-dependent 1δ (ppm1d), also known as wip1 (wild type p53 induced protein phosphatase 1), is a member of the PP2C family of Ser/Thr protein phosphatases. Ppm1d could inhibit p53 signaling, and is putatively oncogenic, . In addition to p53 inhibition, ppm1d down-regulates p38 mitogen-activated protein kinase, . The ppm1d protein is encoded by PPM1D gene, which maps to 17q23.2 and is oncogenic. PPM1D amplification is found in several solid tumors, including medulloblastoma, neuroblastoma, pancreatic adenocarcinoma, ovarian clear cell carcinoma and breast cancer, , , , . For breast cancer, ovarian cancer and lung adenocarcinoma, PPM1D overexpression is associated with poor survival, , .
In this study, we examined PPM1D mRNA expression in HCC and paired non-cancerous liver tissues of 86 HCC patients receiving surgical resection. Potential correlation between PPM1D mRNA expression with clinicopathological characteristics of HCC (e.g., TNM staging and survival) was also analyzed. And the prognostic value of PPM1D mRNA for HCC was investigated.
Materials and Methods
The study protocol was approved by the Ethics Committee of Provincial Hospital Affiliated to Shandong University. All the participants provided their written informed consent for inclusion in the data analysis and manuscript publication..
The current study included 86 HCC patients receiving partial hepatectomy in the Department of Liver Transplantation and Hepatobiliary Surgery, Provincial Hospital Affiliated to Shandong University, during the period from September 2006 to January 2009. The HCC diagnosis was confirmed by post-operative pathological examination. We recorded clinical variables including age, gender, α-FP level (cut-off: 400 ng/L), tumor size (≧5 cm or <5 cm), portal vein invasion, lymph node metastasis, alcohol intake, TNM stage (TNM 3 and 4 as advanced stage), family HCC history (in the parents), recurrence, HBV infection, HCV infection, history of aflatoxin B1 exposure and patient survival time. Samples were collected in the operating room, and stored in liquid nitrogen until use. Non-cancerous liver tissue was defined as liver tissue 5 cm or more away from the tumor border. Both tumor (HCC in nature) and non-cancerous liver tissue control were verified by a pathologist blinded to the tissue origin using hematoxylin-eosin (H&E) staining.
RNA extraction and first strand cDNA synthesis
Total RNA was extracted from frozen tissue samples (30 mg) using AllPrep DNA/RNA Mini Kit (Qiagen, Valencia, CA, USA) according to the manufacturer’s instruction. Genomic DNA was removed using DNA-free™ Kit (Applied Biosystems, Carlsbad, CA, USA) following the protocol. RNA integrity was verified using electrophoresis with 5% agarose/formaldehyde/MOPS(3-(N-Morpholino) propanesulfonic acid) gels followed by ethidium bromide staining and visual inspection under UV light. Samples with the 28S:18S rRNA ratio less than 2∶1 were excluded. The RNA concentration was measured by Nanodrop ND-2000 spectrophotometer (Thermo Fisher Scientific Inc., Waltham, MA). RNA purity was examined by calculating the ratio of absorbance at 260 nm vs. 280 nm. A ratio of ∼2.0 was used as the criterion for “pure” RNA. The quality and integrity of RNA were similar between HCC and non-cancerous liver tissue samples. First strand cDNA was synthesized from 1 µg total RNA, using High Capacity RNA-to-cDNA Kit (Applied Biosystems, Carlsbad, CA, USA) according to the manufacturer’s protocol and under the following conditions: 37°C for 60 min and 95°C for 5 min. First strand cDNA was stored at −20°C until further analysis.
Quantitative real-time polymerase chain reaction
Relative level of PPM1D mRNA in HCC and adjacent non-cancerous liver tissue samples were determined by qPCR. β-ACTIN was used as the internal loading control. qPCR were performed on an ABI StepOne Plus Real-Time PCR System (Applied Biosystems). The primer sequences were: PPM1D: forward, 5'-CAA TTG GCC TTG TGC CTA CT-3'; reverse, 5'- TCT TTC GCT GTG AGG TTG TG -3'. β-ACTIN: forward, 5'-GGA CTT CGA GCA AGA GAT GG-3'; reverse, 5'-AGC ACT GTG TTG GCG TAC AG-3'. All samples were run in triplicate. qPCR amplification was conducted in 20-µL reaction buffer using ABI Power SYBR®Green PCR Master Mix (Applied Biosystems) under the following conditions: 95°C for 10 min, 40 cycles at 95°C for 15 s, and 60°C for 1 min. The melting curve was analyzed for each sample. The amplification products were separated using electrophoresis on 2% agarose gels and visualized by ethidium bromide staining. The expected size of PPM1D is 237bp. The threshold cycle (Ct) was measured in the exponential amplification phase. The amplification plots were analyzed by StepOne v2.2 Software (Applied Biosystems). The Ct values of β-ACTIN was similar across HCC and non-cancerous liver tissues. The results were normalized against β-ACTIN and were expressed as 2-△△Ct.
5-µm-thick paraffin-embedded tissue sections were deparaffinized with xylene and rehydrated with a graded series of ethanol. Endogenous peroxidase was blocked with 0.3% H2O2. Antigen retrieval was performed in 0.1 M sodium citrate buffer (pH∶6.0) with a microwave. Samples were incubated with a rabbit polyclonal antibody against human ppm1d (ab31270, 1∶1000, Abcam, Cambridge, UK) at room temperature and detected using an horseradish peroxidase (HRP) conjugated compact polymer system. Diaminobenzidine (DAB) was used as the chromogen. Slides were counterstained with hematoxylin and mounted with depex. Photographs of immunohistochemical stained sections were taken by a camera mounted on a Keyence BZ-8000 digital microscope (Keyence, Osaka, Japan). Immunochemical staining was examined by two pathologists blinded to the origin of the sections independently.
Evaluation of immunohistochemistry
A semi-quantitative immunoreactivity score (IRS) was used to assess the staining of ppm1d in HCC and non-cancerous liver tissues. Briefly, the intensity was defined as 1, no staining; 2, weak staining; 3, moderate staining; and 4, strong staining. The scoring for the percent of positive cells was defined as 0 (none), 1 (<10%), 2 (10–50%), 3 (51–80%), and 4 (>80%). The scores of intensity and percent of positive cells were multiplied to result in an IRS ranging from 0 to 12 for each specimen. The assessments were carried out by two pathologists independently.
Patient stratification and follow-up
Patients were stratified based on PPM1D mRNA expression using average level (0.24?0.02) in non-cancerous liver tissue samples as cut-off. The follow-up data were obtained from the medical records and direct communication with the patients or their relatives. The follow-up period was defined as the time from the date of surgery to the date of patient death or the last follow-up in August 2011.
All statistical analysis was carried out using SPSS version 13.0 for Windows (SPSS Inc, IL, USA). Quantitative values were presented as mean±standard deviation (SD). The two-tailed Student’s t-test was used to analyze the difference of PPM1D mRNA expression between HCC and non-cancerous liver tissues. And the differential expression of ppm1d protein was assessed by the Wilcoxon test. Correlation analysis between clinical variables and PPM1D mRNA expression was carried out using Fisher’s exact and χ2 tests. The distribution curves of survival time of high expression group and low expression group were analyzed using the Kaplan-Meier method; a log-rank test was used to compute differences between the curves. Prognostic value of PPM1D mRNA expression was examined with a multivariate analysis using the Cox proportional hazards regression model. Statistical significance was set at p value less than 0.05.
The median age of the 86 patients was 48.5 years (range: 24 to 73 years). Seventy-eight subjects were men (78/86, 90.7%), and the remaining 8 (8/86, 9.3%) were women. None of the subjects had HCV infection or history of aflatoxin B1 exposure. HBV infection was found in 69 patients (69/86, 80.2%). HCC had developed in alcohol-related cirrhosis in 38 patients (38/86, 44.2%). No patient was lost to the follow-up. All deaths were associated with HCC recurrence.
Expression of PPM1D mRNA in HCC and non-cancerous liver tissues
The PPM1D mRNA level was higher in HCC tissues than in non-cancerous liver tissues (the ratio against β-ACTIN mRNA at 4.00±0.84 vs. 0.24±0.02; p<0.01; Figure 1A). A representative gel image of PPM1D mRNA amplifications in HCC and non-cancerous liver tissues was presented in Figure 1B. Based on PPM1D mRNA expression, 86 patients were divided into a high expression group (n = 56) and a low expression group (n = 30), using the average level (0.24±0.02) of PPM1D mRNA in non-cancerous liver tissues as cut-off.
(A) PPM1D mRNA expression was examined in 86 pairs of HCC and matched non-cancerous liver tissues with qPCR assay. Data were presented as the abundance relative to β-ACTIN mRNA. PPM1D mRNA in HCC tissues was significantly higher than in the non-cancerous liver tissues (p<0.01). (B) Representative image of 2% agarose gel electrophoresis of qPCR products, the size of PPM1D mRNA product was 237bp. M, marker; C, negative control; T, HCC sample; N, non-cancerous liver sample.
Verification of ppm1d protein expression with immunohistochemistry
Immunohistochemical staining was used to confirm ppm1d protein expression in the 86 HCC tissues and paired non-cancerous liver tissues. The representative images of ppm1d staining were shown in Figure 2A. Immunohistochemistry revealed that ppm1d protein expression in 83 patients (96.5%) was overlapping with mRNA expression. Ppm1d protein expression was higher in HCC tissues than in non-cancerous liver tissues. The difference was statistically significant (p<0.01, Figure 2B).
(A) Representative images of ppm1d staining in HCC and non-cancerous liver tissues. The positive cells were stained dark brown. (a) Negative ppm1d staining in normal liver tissue. (b) Negative staining of HCC tissue without ppm1d primary antibody. (c) Ppm1d-positive non-cancerous liver tissue. (d) High expression of ppm1d in HCC tissue. Magnification, ×200; Scale bar, 100 µm. (B) The distribution of difference of ppm1d staining in HCC and non-cancerous liver tissues (ΔIRS = IRSHCC-IRSnon-cancerous). p<0.01. IRS, immunoreactivity score.
Correlation between PPM1D mRNA expression and clinicopathological features
High expression of PPM1D mRNA in HCC samples correlated with high α-FP level (≥400 ng/L), larger tumor diameters (≥5 cm) and more advanced TNM stage (TNM stage 3 and 4). 82% of the the patients (46/56) in the high expression group had high α-FP level, whereas only 27% of the patients (8/30) in low expression group had high α-FP level. The tumor diameter was more than 5 cm in 59% of the patients (47/56) in high expression group and only 23% of the patients (7/30) in low expression group. 84% of the patients (47/56) in high expression group were in advanced TNM stage, whereas only 37% of the patients (11/30) in low expression group had advanced TNM stage HCC. High PPM1D mRNA expression was associated with family history of HCC. The differences between high expression group and low expression group were statistically significant in terms of these clinical characteristics (α-FP level, p<0.01; tumor diameter, p<0.01; TNM stage, p<0.01 and family history of HCC, p<0.01). The difference of recurrence incidence was statistically significant between the two groups (p<0.01). But no significant differences were found between high expression group and low expression group in terms of age (p = 0.43), gender (p = 1.00), portal vein invasion (p = 0.18), lymphnode metastasis (p = 0.06) or alcohol intake (p = 0.50). There was no significant difference between the two groups in terms of HBV infection (p = 1.00). 45 patients (45/86, 52.3%) in high expression group suffered HBV, and 24 patients (24/86, 27.9%) in low expression group had HBV infection (Table 1).
PPM1D mRNA overexpression correlated with poor prognosis
Kaplan-Meier survival curve and log-rank test were used to analyze the correlation between PPM1D mRNA expression and overall survival. High expression of PPM1D mRNA was associated with poor prognosis (Figure 3). The survival time for high expression group was 13 months (median survival time). That was shorter than the survival time for low expression group (median survival time, 32 months). The difference was statistically significant (p<0.01, hazard ratio; 2.12, 95% CI; 1.22–3.67).
Overall survival was compared between patients with PPM1D mRNA high expression versus low expression with the Kaplan-Meier survival curve. The overall survival in subjects with high PPM1D mRNA expression was significantly lower than in patients with low expression (p<0.01, log-rank test).
The results from the current study showed that PPM1D expression in HCC tissues was significantly higher than in paired non-cancerous liver tissues, at both mRNA and protein levels. This finding is consistent with previous studies that revealed overexpression of PPM1D in other malignant tumors , , .
Hepatocarcinogenesis is a complex process that involves accumulation of genetic and epigenetic changes. Abnormal activation of tumor suppressor genes such as p53 have also been well documented in development of HCC. A variety of risk factors, most notably chronic HBV infection, have been identified for HCC. HBV infection has been implicated in causing HCC through interaction with tumor suppressor genes such as p53. PPM1D is an oncogene. Overexpression of PPM1D contributes to the development of human cancers by suppressing p53 activation. We hypothesized that HBV infection was associated with PPM1D overexpression. To elucidate the hypothesis, we analyzed the correlation between HBV infection and PPM1D mRNA expression. But the current study failed to show significant difference in the HBV infection between the high and low PPM1D mRNA expression groups. However, high PPM1D mRNA expression was associated with family history of HCC. HBV DNA integration is frequently detected in HBV positive HCC cells, and considered to be an important contributing factor in the hepatocarcinogenesis. Murakami et al. demonstrated that HBV integration could lead to aberrant target gene transcription and cause hepatocarcinogenesis. Whether PPM1D overexpression in patients with family history of HCC is associated with HBV integration needs further investigation.
Overexpression of PPM1D is associated with poor prognosis in patients with lung cancer and ovarian cancer, . The results of the present study showed an association of high PPM1D mRNA expression with more aggressive tumor behaviors, including higher α-FP level, advanced TNM stage and higher recurrence incidence. High α-FP level and advanced TNM stage are clinicopathologic marker for invasiveness and unfavorable prognosis in HCC patients, , . We then analyzed overall survival using the Kaplan-Meier survival curve to determine the correlation between the PPM1D mRNA expression and survival. And the results in the present study revealed an association of high PPM1D mRNA expression with shorter survival time. The results proved that PPM1D mRNA expression might be a potential prognostic marker for HCC.
Vascular invasion and lymph node metastasis are also associated with malignant behavior of HCC. However, our study failed to identify a link between PPM1D mRNA expression with vascular invasion or lymph node metastasis. We also failed to find any differences between the two groups in term of other clinical variables including gender, age and alcohol intake. Whether this is caused by the limited sample size needs further studies.
In summary, the current study revealed that PPM1D mRNA was overexpressed in HCC tissues compared with non-cancerous liver tissues and the high PPM1D mRNA expression was an indicator of poor prognosis for HCC patients. Further study is needed to investigate the underlying molecular mechanism of PPM1D in the progression of HCC.
We appreciated Dr. Yu Liu and Lin Han for their help in examing the H&E stained sections and assessing the immunohistochemistry staining.
Conceived and designed the experiments: G-BL JL. Performed the experiments: G-BL X-LZ LY Q-QJ. Analyzed the data: G-BL D-JL JL. Contributed reagents/materials/analysis tools: X-LZ LY Q-QJ. Wrote the paper: G-BL X-LZ JL.
- 1. Xu L, Qian G, Tang L, Su J, Wang JS (2010) Genetic variations of hepatitis B virus and serum aflatoxin-lysine adduct on high risk of hepatocellular carcinoma in Southern Guangxi, China. J Hepatol 53: 671–676.
- 2. Kishi Y, Hasegawa K, Sugawara Y, Kokudo N (2011) Hepatocellular carcinoma: current management and future development-improved outcomes with surgical resection. Int J Hepatol 2011: 728103.
- 3. Hussain SP, Schwank J, Staib F, Wang XW, Harris CC (2007) TP53 mutations and hepatocellular carcinoma: insights into the etiology and pathogenesis of liver cancer. Oncogene 26: 2166–2176.
- 4. Staib F, Hussain SP, Hofseth LJ, Wang XW, Harris CC (2003) TP53 and liver carcinogenesis. Hum Mutat 21: 201–216.
- 5. Mazzanti R, Gramantieri L, Bolondi L (2008) Hepatocellular carcinoma: epidemiology and clinical aspects. Mol Aspects Med 29: 130–143.
- 6. Lim SO, Park YM, Kim HS, Quan X, Yoo JE, et al. (2011) Notch1 differentially regulates oncogenesis by wildtype p53 overexpression and p53 mutation in grade III hepatocellular carcinoma. Hepatology 53: 1352–1362.
- 7. Lu X, Nguyen TA, Moon SH, Darlington Y, Sommer M, et al. (2008) The type 2C phosphatase Wip1: an oncogenic regulator of tumor suppressor and DNA damage response pathways. Cancer Metastasis Rev 27: 123–135.
- 8. Shreeram S, Hee WK, Demidov ON, Kek C, Yamaguchi H, et al. (2006) Regulation of ATM/p53-dependent suppression of myc-induced lymphomas by Wip1 phosphatase. J Exp Med 203: 2793–2799.
- 9. Bulavin DV, Demidov ON, Saito S, Kauraniemi P, Phillips C, et al. (2002) Amplification of PPM1D in human tumors abrogates p53 tumor-suppressor activity. Nat Genet 31: 210–215.
- 10. Koom WS, Park SY, Kim W, Kim M, Kim JS, et al. (2012) Combination of radiotherapy and adenovirus-mediated p53 gene therapy for MDM2-overexpressing hepatocellular carcinoma. J Radiat Res 53: 202–210.
- 11. Bulavin DV, Phillips C, Nannenga B, Timofeev O, Donehower LA, et al. (2004) Inactivation of the Wip1 phosphatase inhibits mammary tumorigenesis through p38 MAPK-mediated activation of the p16(Ink4a)-p19(Arf) pathway. Nat Genet 36: 343–350.
- 12. Castellino RC, De Bortoli M, Lu X, Moon SH, Nguyen TA, et al. (2008) Medulloblastomas overexpress the p53-inactivating oncogene WIP1/PPM1D. J Neurooncol 86: 245–256.
- 13. Saito-Ohara F, Imoto I, Inoue J, Hosoi H, Nakagawara A, et al. (2003) PPM1D is a potential target for 17q gain in neuroblastoma. Cancer Res 63: 1876–1883.
- 14. Loukopoulos P, Shibata T, Katoh H, Kokubu A, Sakamoto M, et al. (2007) Genome-wide array-based comparative genomic hybridization analysis of pancreatic adenocarcinoma: identification of genetic indicators that predict patient outcome. Cancer Sci 98: 392–400.
- 15. Tan DS, Lambros MB, Rayter S, Natrajan R, Vatcheva R, et al. (2009) PPM1D is a potential therapeutic target in ovarian clear cell carcinomas. Clin Cancer Res 15: 2269–2280.
- 16. Yu E, Ahn YS, Jang SJ, Kim MJ, Yoon HS, et al. (2007) Overexpression of the wip1 gene abrogates the p38 MAPK/p53/Wip1 pathway and silences p16 expression in human breast cancers. Breast Cancer Res Treat 101: 269–278.
- 17. Rauta J, Alarmo EL, Kauraniemi P, Karhu R, Kuukasjarvi T, et al. (2006) The serine-threonine protein phosphatase PPM1D is frequently activated through amplification in aggressive primary breast tumours. Breast Cancer Res Treat 95: 257–263.
- 18. Satoh N, Maniwa Y, Bermudez VP, Nishimura K, Nishio W, et al. (2011) Oncogenic phosphatase Wip1 is a novel prognostic marker for lung adenocarcinoma patient survival. Cancer Sci 102: 1101–1106.
- 19. Weichert W, Roske A, Gekeler V, Beckers T, Ebert MP, et al. (2008) Association of patterns of class I histone deacetylase expression with patient prognosis in gastric cancer: a retrospective analysis. Lancet Oncol 9: 139–148.
- 20. Aravalli RN, Steer CJ, Cressman EN (2008) Molecular mechanisms of hepatocellular carcinoma. Hepatology 48: 2047–2063.
- 21. Um TH, Kim H, Oh BK, Kim MS, Kim KS, et al. (2011) Aberrant CpG island hypermethylation in dysplastic nodules and early HCC of hepatitis B virus-related human multistep hepatocarcinogenesis. J Hepatol 54: 939–947.
- 22. Jiang S, Yang Z, Li W, Li X, Wang Y, et al. (2012) Re-evaluation of the carcinogenic significance of hepatitis B virus integration in hepatocarcinogenesis. PLoS One 7: e40363.
- 23. Murakami Y, Saigo K, Takashima H, Minami M, Okanoue T, et al. (2005) Large scaled analysis of hepatitis B virus (HBV) DNA integration in HBV related hepatocellular carcinomas. Gut 54: 1162–1168.
- 24. Ali AY, Abedini MR, Tsang BK (2012) The oncogenic phosphatase PPM1D confers cisplatin resistance in ovarian carcinoma cells by attenuating checkpoint kinase 1 and p53 activation. Oncogene 31: 2175–2186.
- 25. Peng SY, Chen WJ, Lai PL, Jeng YM, Sheu JC, et al. (2004) High alpha-fetoprotein level correlates with high stage, early recurrence and poor prognosis of hepatocellular carcinoma: significance of hepatitis virus infection, age, p53 and beta-catenin mutations. Int J Cancer 112: 44–50.
- 26. Yen YH, Changchien CS, Wang JH, Kee KM, Hung CH, et al. (2009) A modified TNM-based Japan Integrated Score combined with AFP level may serve as a better staging system for early-stage predominant hepatocellular carcinoma patients. Dig Liver Dis 41: 431–441.
- 27. Minagawa M, Ikai I, Matsuyama Y, Yamaoka Y, Makuuchi M (2007) Staging of hepatocellular carcinoma: assessment of the Japanese TNM and AJCC/UICC TNM systems in a cohort of 13,772 patients in Japan. Ann Surg 245: 909–922.
- 28. Lin SY, Pan HW, Liu SH, Jeng YM, Hu FC, et al. (2008) ASPM is a novel marker for vascular invasion, early recurrence, and poor prognosis of hepatocellular carcinoma. Clin Cancer Res 14: 4814–4820.