Brief Reports
Copyright ©The Author(s) 2000. Published by Baishideng Publishing Group Inc. All rights reserved.
World J Gastroenterol. Feb 15, 2000; 6(1): 134-135
Published online Feb 15, 2000. doi: 10.3748/wjg.v6.i1.134
Overexpression of P-glycoprotein in hepatocellular carcinoma and its clinical implication
Xian-Bing Kong, Zu-Kui Yang, Li-Jian Liang, Jie-Fu Huang, Han-Liang Lin
Xian-Bing Kong, Zu-Kui Yang, Department of Surgery, The First Affiliated Hospital, Chongqing University of Medical Sciences, Chongqing 400016, China
Li-Jian Liang, Jie-Fu Huang, Division of Hepatobiliary Surgery, The First Affiliated Hospital, Sun Yat-Sen University of Medical Sciences, Guangzhou 510080, Guangdong Province, China
Han-Liang Lin, Department of Pathology, Sun Yat-Sen University of Medical Sciences, Guangzhou 510089, Guangdong Province, China
Xian-Bing Kong, male, born in 1961-10-17 in Pengxi County, Sichuan Province, graduated from Chongqing University of Medical Sciences as postgraduate in 1989, having 8 papers published.
Author contributions: All authors contributed equally to the work.
Correspondence to: Dr. Xian-Bing Kong, Department of Surgery, The First Affiliated Hospital, Chongqing University of Medical Sciences, Chongqing 400016, China
Telephone: +86-23-68899866
Received: June 23, 1999
Revised: August 22, 1999
Accepted: September 18, 1999
Published online: February 15, 2000

Key Words: liver neoplasms, carcinoma, hepatocellular, P-glycoprotein, multidrug resistance-1 gene, immunohistochemistry


Most advanced hepatocellular garcinoma (HCC) is insensitive to most anticancer drugs which might be related to the high frequency of expression of the multidrug resistance-1 (MDR1) gene[1] and its product, P-glycoprotein (P-gp)[2]. P-gp expression may also be concerned with tumor progression and differentiation[3]. In the present study, we investgated P-gp expression and assessed the relationship between expression level of P-gp and the clinico-pathological parameters of HCC by immunohistochemistry in combination with computer-imaging analysis.

Patients and specimens

The data were from 47 HCC resected specimens, from June 1996 to July 1997, including 47 tumor tissues and 33 adjacent tissues, 39 men and 8 women, age 25-74 years (mean 49 years). No one received preoperative chemotherapy. Of the 47 tumors, 72% were > 5 cm, and 28% were ≤ 5 cm, 40% and 26% were accompanied by involvement of portal veins with presence of satellite nodules and intrahepatic metastasis, respectively. The HCC consisted of well and poorly differentiated types and 11 normal liver tissues were obtained at surgery from patients without chronic liver disease as control.

Immunohistochemical stainings for P-gp

Serial sections of 4ìm in thickness were prepared and immunohistochemical examination was performed by the streptavidin- biotin-complex (SABC) method using monoclonal antibody JSB1 (Boehringer Mannheim). The negative control was processed by substituting PBS for the primary antibody. The positive control was a breast cancer sample shown to express P-gp by SABC method.

Quantitative analysis

Immunoprecipitates were analyzed using an Axiotron microscope (Zeiss, Germany) and a 3CCD camera (JVC, Japan), and then processed by an image analysis system (IBAS, Kontron, Germany). According to Shen’s method[4], only the gray level and the area of defined area A and the phase á of positive products in A need to be tested. Positive unit (PU) was tested and calculated with the formula in A: PU = 100 × |Gα - GA / (1 - A) × Gmax. Gα and GA denote respectively the mean gray level of the studied structure α and the test area A. AAα means the area density of phase α in A. Gmax is the maximum gray level of the instrument. PU that reflected intensity and area of the phase α of positive products in A represented P-gp ex pression level. Five times were performed at random in a section to obtain the mean PU.

Statistical analysis

PU of P-gp was expressed as mean ± SD. The data were analyzed statistically by means of Student’s t test. Statistical significance was defined as a two-sided P value of P < 0.05.

Positive expression of P-gp in HCC, adjacent to tumor and normal liver tissues

P-gppositive cells were not uniformly distributed, the intensity of immunoreactivity was also variable, staining was mainly on the cellular membrane. In normal liver, P-gp was present on the bile canalicular surface of hepatocytes and the luminal surface of bile duct epithelial cells. In P-gp-positive HCC, the staining was observed on the contact surface between tumor cells resembling the canalicular staining pattern. P-gp expression was found in 70% (33/47) of HCC tissues and 66% (22/33) of adjacent tissues. The staining was mild in three normal liver tissues specimens. A significant increase in P-gp expression level was observed in HCC tissues (P < 0.01) and those adjacent to the tumor (P < 0.01) but without significant difference (P > 0.05).

P-gp expression level in relation to clinico-pathological parameters in HCC

The expression level of P-gp in HCC with local invasion was significantly higher than that in HCC without (P < 0.05). Moreover, it was significantly higher in HCC of the well differentiated type than that of poorly differentiated (P < 0.01). Recurrence of HCC also had a significantly higher P-gp expression than that without (P < 0.01). On the other hand, there were no relationships between P-gp expression level and the tumor size (P > 0.05).


Hepatocytes exclusively express P-gp on the bile canalicular surface, it is a Mr170000 membrane protein responsible for pumping lipophilic anticancer drugs such as doxorubicine and vincristine out of tumor cells[5], resulting in chemotherapeutic effect poor. Recent reports showed increased expression of P-gp in HCC was significantly associated with non-responders[6]. Our results indicated more than two thirds of the samples were positive for P-gp in HCC and adjacent tissues, this suggested that P-gp overexpression be probably responsible for the so-called innate drug resistance of HCC or non responder.

The fact that P-gp expression level was significantly higher in well differentiated than poorly differentiated HCC ( P < 0.01 ), suggested the well differentiated HCC have a stronger potential of P-gp expression and a tendency to produce resistance to anticancer drugs more easily. It is worth trial to treat with anticancer drugs in combination with a reverse agents of MDR1 gene or P-gp modulator as verapamil, preferably supported by data on P-gp expression.

The intrahepatic spreading, metastasis and postoperative recurrence were closely related to the biologic behavior of tumor cells, etc, about 43.5% of patients with small HCC recurred within 5 years after surgical resection[7]. Recently, P-gp overexpression had been demonstrated in cancer with recurrence[8]. So far there have been no reports on P-gp expression in HCC recurrence. In this study, hepatic recurrence had a significantly higher P-gp expression level than that without (P < 0.01). These results revealed that recurrence of hepatoma probably more easily to acquire multidrug resistance, thereby, the response to chemotherapy was likely unsatisfactory.

Our study further showed P-gp expression level of invasive HCC was significantly higher than that of non-invasive HCC(P < 0.05), suggesting that a high P-gp expression level was well associated with the invasiveness of HCC, and not related to the size of the tumor whether large or small.


Supported in part by the China Medical Board of New York, Inc., USA, Grant No.90-534

Edited by Wu XN

Proofread by Miao QH

1.  Gottesman MM. How cancer cells evade chemotherapy: sixteenth Richard and Hinda Rosenthal Foundation Award Lecture. Cancer Res. 1993;53:747-754.  [PubMed]  [DOI]
2.  Park JG, Lee SK, Hong IG, Kim HS, Lim KH, Choe KJ, Kim WH, Kim YI, Tsuruo T, Gottesman MM. MDR1 gene expression: its effect on drug resistance to doxorubicin in human hepatocellular carcinoma cell lines. J Natl Cancer Inst. 1994;86:700-705.  [PubMed]  [DOI]
3.  Chin KV, Ueda K, Pastan I, Gottesman MM. Modulation of activity of the promoter of the human MDR1 gene by Ras and p53. Science. 1992;255:459-462.  [PubMed]  [DOI]
4.  Shen H. Study on the quantitative method of immunohistochem-istry (3). Chin J Histocytochem. 1995;4:89-91.  [PubMed]  [DOI]
5.  Goldstein LJ. MDR1 gene expression in solid tumours. Eur J Cancer. 1996;32A:1039-1050.  [PubMed]  [DOI]
6.  Chou YY, Cheng AL, Hsu HC. Expression of P-glycoprotein and p53 in advanced hepatocellular carcinoma treated by single agent chemotherapy: clinical correlation. J Gastroenterol Hepatol. 1997;12:569-575.  [PubMed]  [DOI]
7.  Tang ZY. Studies on small hepatocellular carcinoma clinical aspect and molecular biology. Hepatol Rapid Literat Rev. 1996;26:11-19.  [PubMed]  [DOI]
8.  Gregorcyk S, Kang Y, Brandt D, Kolm P, Singer G, Perry RR. p-Glycoprotein expression as a predictor of breast cancer recurrence. Ann Surg Oncol. 1996;3:8-14.  [PubMed]  [DOI]