ROLE OF SERUM GLYPICAN-3 (GPC-3) IN EARLY DETECTION OF HEPATOCELLULAR CARCENOMA

By

Mohamed M. Hashem*,  Nabil Salma Mohmuod**, 

Gamal Mohamed Soliman**, Mohamed Ahmed Atallah***

and Olfat M. Hendy****

Departments of Internal Medicine* and Tropical Medicine**, Al Azhar University, ClinicalPathology,

Institute of Tropical Medicine and Hepatology*** and  Department  of National Liver Institute,

Monoufyia University****.

ABSTRACT

Background The false negative rate of alpha-fetoprotein (AFP) level alone may be as high as 40% for patients with early stage of hepatocellular carcinoma (HCC) and may be elevated in non-malignant chronic liver diseases and other malignancies. Glypican-3 (GPC-3) is a new tumor marker for HCC.

Objective: Evaluation of the role of serum GPC-3 in the early diagnosis of HCC.

Results: There was a highly significant difference between control and HCC group as regard AFP and serum GPC-3. There  was a highly  significant difference between liver cirrhosis and HCC group as regard AFP and serum GPC-3. No significant difference between control and liver cirrhosis groups as regard AFP and GPC-3. AFP showed sensitivity (75%), specificity (63%), positive predictive value (PPV) (62.8%) and negative predictive value (NPV) (69.9%) at cut-off value 195ng/ml for HCC group. Serum GPC-3 showed sensitivity of (87%), specificity of (95%), PPV of (93.8%) and NPV of (91.2%) at cut-off 5.1ng/ml for HCC group. Combined serum GPC-3 and AFP showed sensitivity (85.9%), specificity (88%) PPV (85.4%) and NPV (84.1%) at cut of value 5.1 ng/ml and 195 ng/ml respectively for HCC group.No significant correlation of serum GPC- 3 or AFP to the tumor size, number or vascular invasion in HCC group.

Conclusion:GPC-3 is a promising diagnostic marker with high sensitivity and specificity for HCC than AFP in detection, screening HCC and follow up treatment of HCC.

INTRODUCTION

     Hepatocellular carcinoma (HCC) is the fifth most common cancerworldwide and the third most common cause of cancer mortality (Lai. et al., 2002 and El-Serag & Rudolph, 2007). AFP has been used as a serum marker for HCC in humans for many years and it has a sensitivity of 39%- 65%, a specificity of 76%-94% and apositive predictive value of 9% - 50% according to the cutoff level used (Daniele et al., 2004).

Not all HCC secrete AFP. Only a small proportion of tumors at an early stage (10-20%) present with abnormal AFP serum level (Yamashita et al., 2008). Serum AFP level may be elevated in non-malignant chronic liver diseases including chronic hepatitis and cirrhosis, as well as in other primary and secondary liver cancers. Therefore, the identification of alternative serum markers of HCC is needed (Spangenberg et al., 2006).

      Glypican-3 (GPC-3) belongs to the glypican family of glycosyl-phosphatidy-linositol (GPI)-anchored heparan sulfate proteoglycans, which plays an important role in cellular growth, differentiation and migration (Filmus and Selleck, 2001). GPC-3 as a potential tumor marker for HCC was first suggested by Hsu et al., (1997). GPC3 has been reported to be increased in HCC in comparison with pre-neoplastic lesions and cirrhotic tissues at the mRNA and protein levels
( Liovet et al., 2006 ).

      In contrast, GPC-3 has been shown to be down-regulated in various cancers including breast cancer, ovarian cancer, lung adenocarcinoma and cholangiocarcinoma (Man et al., 2006). Given the detection of circulating GPC-3, it appears that thisoncofetal protein can serve as a potential serum marker for the detection ofHCC. However, data on its clinical correlations in patients with HCC arecurrently unknown (Hippo et al., 2004).

     The present work aimed to evaluate the role of serum GPC-3 in the early detection of HCC.  

PATIENTS AND METHODS

     In this prospective study, we randomly selected 70 patient 40 of them have proved HCC (29 males, 11 females), and 30 patient with liver cirrhosis (21 males and 9 females). This in addition to 20 healthy subjects (15 males and 5 females). Age ranged between 55.7 ± 11.9 (26 and 76). They were selected from Outpatient clinic and Inpatient Department of Tropical Medicine in Al-Hussein and Sayed Galal University Hospitals, Hepatology Department - National Institute - Monufiya University and Mabara Insurance Hospital (Assiut). The study was done from January 2011 to May 2014. The study was approved by the local ethical committee in Al-Azhar University Hospitals and informed consent was obtained from the patients.

I. HCC group: The patients were diagnosed as HCC according to American Association for study of liver disease (Bruix and Sherman, 2011).

None of the patients had received local or systemic therapy for HCC before. Any patient with cancer other than HCC was excluded.

II. Liver cirrhosis group: Patients were selected according to clinical examination, abdominal U/S, labora-tory investigation and histopathologi-cal examination in liver biopsy samples. Patients were diagnosed to have liver cirrhosis if they have the clinical splenomegaly, ascites, thrombocytopenia, decreased S. albumin, increased bilirubin, prolonged prothrombin time and endoscopic examination (varices).

III. Control group: Subjectswereselec-ted according to clinical examination, abdominal U/S, and laboratory investigation.

     Each patients and control were subjected to Careful history and thorough clinical examination, liver function tests, PT, PC, INR, HCV Ab and HBsAg, CBC, AFP and GPC-3, abdominal U/S: All patients in the HCC group were diagnosed by triphasic spiral CT abdomen and/or dynamic MRI. One patient of HCC group was diagnosed with liver biopsy.

Statistical analysis of data was carried out using the SPSS computer package version 17.0 (SPSS Inc. Chicago. IL, USA). The mean ± SD was used for quantitative variables while number and% were used for qualitative variables. Independent samples t-test was applied. In order to assess the differences in means of quantitative variables within the patients group. One-Way ANOVA test was used. Correlation analysis (using Pearson's correlation coefficient) was used to assess the strength of association between various quantitative variables within the patients' group. The statisti­cal methods were verified, assuming a significance level of p < 0.05. Receiver operating characteristic curve (ROC) was performed on SPSS-16 to determine the sensitivity and specificity of serum GPC-3 and AFP.

RESULTS

      No significant difference between control and liver cirrhosis as regard AFP and serum GPC-3, while there was significant difference between control and HCC. A highly significant difference was found between liver cirrhosis and HCC groups (Table 1).

Table (1): Comparison between control, liver cirrhosis and HCC groups regarding AFP, GPC-3..

Fifteen cases (37.5%) of HCC group have tumor size ≤ 5cm or 3 lesions each ≤ 3 cm. withno significant difference between groups. Twenty nine cases of HCC group (72.5%) have a single lesion, withno significant differences between tumor number and GPC-3 or AFP (Table 2).

Table (2): Correlation of tumor size and number in HCC group to GPC-3 and AFP.

     Twelve cases (30%) of HCC group showed vascular invasion. No significant correlation was found between serum GPC-3 or AFP levels and vascular invasion (Table 3).

Table (3):Correlation of vascular invasion and Serum GPC-3 in HCC group.

      On differentiating HCC from cirrhosis and normal controls, AFP showed sensitivity 75%, specificity 63%, Positive Predictive Value (PPV) 62.8% and Negative Predictive Value (NPV) 69.9% at cut-off value 195ng/ml. Serum GPC-3 showed sensitivity of 87%, specificity of 95%, PPV 93.8% and NPV 91.2% at cut-off 5.1ng/ml. Combined serum GPC-3 and AFP showed sensitivity 85.9% specificity 88% PPV 85.4% and NPV 84.1% at cut of value 5.1 ng/ml and 195 ng/ml respectively (Table 4).

Table (4): Sensitivity, specificity, PPV and NPV of serum GPC-3 and AFP in prediction of HCC.

DISCUSSION

        HCC is the most common primary malignant tumor of the liver, accounts for >80% of liver cancer cases, and affects men 3 to 4 times more frequently than women. The burden of HCC has been increasing in Egypt with a doubling in the incidence rate in the past 10 years (Mizokami and Tanaka, 2005).

      AFP is an oncofetal proteins produced during fetal life, disappear after birth and reappear in cancer patients (Marrero et al., 2009). AFP levels at value of 20 ng/ml showed good sensitivity but low specificity, whereas, at higher cut-offs of 200 ng/ml sensitivity drops to 22% with high specificity (Trevisani et al., 2001). Furthermore, only a small proportion of tumors at an early stage (10-20%) present with abnormal AFP serum level (Yamashita et al., 2008). The AASLD recommended cut-off level for diagnosis of HCC is200ng/ml. Although lower levels, particularly if rising, should be followed very carefully (Bruix and Sherman, 2011).

      The sensitivity of AFP is not good enough for it to be used alone for detection of HCC as over a third of cancers will be missed (Bruix and Sherman, 2005).

      Normally, GPC-3 plays a role in regulating cell proliferation and survival during embryonic development by modulating the activity of various growth factors. Under normal physiological conditions, the tissue GPC-3 protein resides in the endoplasmic reticulum and is transported into the nucleus under stress (Pang et al.,  2008 and Abd El Gawad, et al., 2014). GPC-3 is only detected in HCC cells, not in benign liver tissues andcan thus be used as a potential biomarker for the screening and diagnosisof early HCC. Serum GPC3 level was higher than 300 ng/l in 50% ofearly HCC patients, although their serum AFP level was below 100 μg/l (Lee et al., 2014).

      Our study aimed to determine serum level of GPC-3 in patients with HCC, liver cirrhosis and healthy control to find its value in relation to serum AFP for early HCC detection. As regard correlation of AFP to serum GPC-3 no significant difference between normal control and liver cirrhosis, but in HCC group a significant positive correlation was detected between AFP and GPC-3 in the HCC group. Our results were in agreement with El-Shenawy et al. (2012). Contrarily, no correlation was found between serum levels of AFP and GPC-3 (Jackbovic and Jothy, 2007).

     As regard relation of serum GPC-3 and AFP to tumor size, number and vascular invasion, our study showed that 37.5% of HCC patients have a single lesion ≤ 5 cm or three lesions each less each ≤ 3 cm. Also, 30% of HCC group have vascular invasion at the time of detection with no significant correlation of serum GPC- 3 or AFP to the tumor size, number or vascular invasion. Ozkan et al. (2011) found no correlation between GPC-3 levels and tumor size, number and portal vein invasion in patients with HCC. A positive correlation was found between serum levels of each of AFP and GPC-3 with both tumor size and portal vein invasion by El- Shenawy et al., (2012), while Youssef et al. (2010) reported statistically significant results between GPC-3 and the staging of HCC. This discrepancy of results may be due to the different sample size or the different underlying etiology of HCC.

      Serum GPC-3, in our study, showed a cut-off level of 5.1ng/ml, gave sensitivity of 87%, specificity of 95%, PPV 93.8% and NPV 91.2% for early detection of HCC and AUROC of 0.198. Nakatsura et al.(2003) stated that GPC-3 could be detected in 40–53% of HCC patients and 33% of AFP sero-negative HCC patients, while Shafizadeh et al. (2008) found GPC3 positive cells in 90% of patients with their serum AFP level < 400 ng/ml.

      As regard AFP, in our study, gave a sensitivity of 75% and specificity of 63%, PPV 62.8% and NPV69.9% at cut-off level of 195 ng/ml and AUROC of 0.266.  The combined AFP and serum GPC-3 gave sensitivity of 85.9%, specificity of 88%, PPV 85.4% and NPV 84.1% which is much better than AFP alone. Combination of GPC-3 and AFP showed that the sensitivities were between 84–92% and specificities between 90–95% (Gomaa et al., 2012). Serum GPC-3 levels in patients with small-sized HCC tumors (60%, while the sensitivity of AFP was extremely low <10% (Lee et al., 2014).

     Hippo et al. (2004) demonstrated that during the follow-up of their cirrhotic patients having detectable GPC-3 levels, HCC developed within 6 months among considerable number of patients with neither significant change of serum AFP levels nor in abdominal U/S.

     AFP proteins do not seem to play a critical role in tumor progression, but have been used as tumor markers or as targets for immunotherapy. Evaluation of GPC-3 as a diagnostic and immunotherapeutic target may be used in the future for treatment of HCC Sung et al. (2003).

     In conclusion, serum GPC-3 is a sensitive and specific tumor marker for early detection and for follow up after treatment of HCC.

REFERENCES

1. Abd El Gawad IA, Mossallam GI, Radwan NH, Elzawahry HM and Elhifnawy NM. (2014): Comparing Prothrombin induced by vitamin K absence-II (PIVKA-II) with the oncofetal proteins GPC-3, AFP and CEA in diagnosing HCC among Egyptian patients. Journal of the Egyptian National Cancer Institute ; 26: 79–85.

2. Bruix J and Sherman M (2005):  Management of HCC. Hepatology, 42: 1208–1236.

3. Bruix, J. and: Sherman, M (2011):  Manage-ment of hepatocellular carcinoma: an update. Hepatology53, 1020–1022.

4. Daniele B, Bencivenga A, Megana AS and Tinessa V. (2004): AFP and U/S screening for HCC. Gastroenterology, 127 (5): S108-S112.

5. El-Serag H and Rudolph K (2007): HCC: Epidemiology and Molecular Carcinogenesis. Gastroenterology, 132(7): 2557-2576.

6. El-Shenawy S, El Sabawi M, Sheble N,  Abd El-Raof M; Allam M and Fath Allah S. (2012): Diagnotic role of serum GPC-3 as a tumor marker for HCC. Nat. Sci., 10(4):32–8.

7. Filmus J and Selleck SB (2001): Glypicans: proteoglycans with a surprise. J. Clin. Invest., 108: 497–501.

8. Gomaa A, Hendy O, Aboraia G, Attia H, El Ezawy H, Nafie E and  El Baz S. (2012): The diagnostic value of peripheral blood Glypican-3 in patients with hepatocellular carcinoma. World J Med Sci., 7(2):105–12.

9. Hippo Y, Watanabe K, Watanabe A,  Midorikawa Y, Yamamoto S, Ihara S, Tokita S, Iwanari H, Ito Y, Nakano K, Nezu J, Tsunoda H, Yoshino T, Ohizumi I,Tsuchiya M, Ohnishi S, Makuuchi M, Hamakubo T, Kodama T and Aburatani H.. (2004): Identification of soluble NH2-terminal fragment of GPC-3 as a serological marker for early-stage HCC. Cancer Res., 64: 2418–2423.

10. Hsu HC, Cheng W and Lai PL (1997): Cloning and expression of a developmentally regulated transcript MXR7 in HCC: biological significance and temporospatial distribution. Cancer Res., 57: 5179–84.

11. Jackbovic B and Jothy S (2007): GPC-3 from the mutations of Simpson-Golabi-Behmel genetic syndrome to a tumor marker for hepatocellular carcinoma. Exp Mol Pathol, 82, 184-9.

12. Lai C.W.Y., Wardle M, Birkett V, Powell V, Thomas G.A.O  (2002)SURVEILLANCE FOR HEPATOCELLULAR CARCINOMA IN CIRRHOSIS: A NATIONAL AUDIT OF THE PRACTICE IN THE UK. Gut, 50:a111-A119. 

13. Lee HJ, Yeon JE, Suh SJ, Lee SJ, Yoon EL, Kang K, Yoo YJ, Kim JH, Seo YS, Yim HJ and Byun KS (2014): Clinical Utility of Plasma GPC-3 and Osteopontin as Biomarkers of HCC. Gut and Liver, 177-185.

14. Liovet JM, Burroughs A and Bruix J. (2006): HCC. Lancet, 362:1907–17.

15. Man XB, Tang L, Zhang BH,  Li SJ, Qiu XH, Wu MC and Wang HY. (2006): Upregulation of GPC-3 expression in HCC but downregulation in cholangiocarcinoma indicates its differential diagnosis value in primary liver cancers. Liver Int., 25: 962–6.

16. Marrero JA1, Feng Z, Wang Y, Nguyen MH, Befeler AS, Roberts LR, Reddy KR, Harnois D, Llovet JM, Normolle D, Dalhgren J, Chia D, Lok AS, Wagner PD,Srivastava S and  Schwartz M (2009): Alpha-fetoprotein, des-gamma carboxyprothrombin, and lectin-bound alpha-fetoprotein in early hepatocellular carcinoma. Gastroenterology, 137(1):110-8.

17. Mizokami M and Tanaka Y (2005): Tracing the evolution of HCV in the US, Japan and Egypt by using the molecular clock. Clin Gastroenterol Hepatol., 3: 82–5

18. Nakatsura T, Yoshitake Y, Senju S,  Monji M, Komori H, Motomura Y, Hosaka S, Beppu T, Ishiko T, Kamohara H, Ashihara H, Katagiri T, Furukawa Y, Fujiyama S, Ogawa M, Nakamura Y and Nishimura Y. (2003): GPC-3, over expressed specifically in human HCC is a novel tumor marker. Biochem Biophys Res Commun., 306 (1):16–25.

19. Ozkan H, Erdal H, Koc¸ak E , Tutkak H, Karaeren Z, Yakut M and Köklü S.. (2011): Diagnostic and prognostic role of serum GPC-3 in patients with HCC. J Clin Lab Anal., 25(5):350–3.

20. Pang RW1, Joh JW, Johnson PJ, Monden M, Pawlik TM and Poon RT.(2008): Biology of hepatocellular carcinoma.Ann Surg Oncol.,  15(4):962-71.

21. Shafizadeh N, Ferrel P and Karkar S (2008): Utility and limitation of GPC-3 expression for HCC at both ends of the differentiation spectrum. Mod Pathol., 21(8):1011–8.

22. Spangenberg HC, Thimme R and Blum HE (2006): Serum markers of HCC. Semin. Liver Dis., 26: 385–90.

23. Sung YK, Hwang SY, Park MK,  Farooq M, Han IS, Bae HI, Kim JC and  Kim M. (2003): GPC-3 is overexpressed in human HCC. Cancer Sci,, 94: 259-62.

24. Trevisani F, D'Intino PE, Morselli-Labate AM, Mazzella G, Accogli E, Caraceni P, Domenicali M, De Notariis S, Roda E and  Bernardi M. (2001): Serum alpha-fetoprotein for diagnosis of hepatocellular carcinoma in patients with chronic liver disease: influence of HBsAg and anti-HCV status.J Hepatol., 34(4):570-5.   

25. Yamashita T, Forgues M, Wang W, Kim JW, Ye Q, Jia H, Budhu A, Zanetti KA, Chen Y, Qin LX, Tang ZY and  Wang XW (2008): EpCAM and AFP expression defines novel prognostic subtypes of HCC. Cancer Res., 68: 1451-1461.

26. Youssef M, El-Sharkawy S and Abbas N. (2010): Clinical utility of GPC-3 in HCC. Int J Integr Biol., 10(1):41–7.