Article Text

Download PDFPDF

Glypican-3 expression predicts poor clinical outcome of patients with early-stage clear cell carcinoma of the ovary
  1. Tomokazu Umezu,
  2. Kiyosumi Shibata,
  3. Hiroaki Kajiyama,
  4. Eiko Yamamoto,
  5. Akihiro Nawa,
  6. Fumitaka Kikkawa
  1. Department of Obstetrics and Gynecology, Nagoya University Graduate School of Medicine, Nagoya, Japan
  1. Correspondence to Tomokazu Umezu, Department of Obstetrics and Gynecology, Nagoya University Graduate School of Medicine, 65 Tsurumai-cho, Showa-ku, Nagoya 466-8550, Japan; t-ume{at}


Background Glypican-3 (GPC3), a membrane-bound heparan sulphate proteoglycan, may play a role in promoting cancer cell growth and differentiation. Recent studies reported that GPC3 is overexpressed in clear cell carcinoma (CCC) of the ovary, and not other ovarian histotypes. However, in CCC patients, the relationship between the overexpression of GPC3 and prognosis has not yet been clarified.

Aim To evaluate GPC3 expression by immunohistochemistry in CCC.

Methods and Results In 52 CCC patients, GPC3 expression was observed in 40.4%. In cases of CCC, no correlations were identified between GPC3 expression and clinicopathological factors, such as age, FIGO stage, CA125 values, peritoneal cytology, ascitic fluid volume and mortality rate, except for the residual tumour size. GPC3 expression was associated with poor progression-free survival in stage I CCC patients. The numbers of Ki-67-stained cells in GPC3-positive areas were lower than those in GPC3-negative areas. GPC3 expression may be associated with a low proliferation rate in CCC cells. In the early stage of CCC, GPC3-expressing patients tended to be resistant to taxane-based treatment.

Conclusions Results suggest that the overexpression of GPC3 may be related to the low-level proliferation of tumours; it may be associated with resistance to taxane-based chemotherapy and a poor prognosis in CCC of the ovary.

  • GPC3
  • immunohistochemistry
  • CCC
  • prognosis
  • early-stage
  • cancer
  • ovary

Statistics from


Epithelial ovarian carcinoma (EOC) is the leading cause of death from gynaecological malignancy. Since ovarian carcinoma frequently remains clinically silent, the majority of patients with this disease have advanced intraperitoneal metastatic disease at diagnosis.1 In addition, various histological types and degrees of malignancy make it complicated to understand and analyse ovarian carcinoma, and the chemosensitivity and biological nature are different among these histological types.2 Clear cell carcinoma (CCC) of the ovary was originally termed ‘mesonephroma’ by Schiller in 1939, as it was thought to originate from mesonephric structures and resemble renal carcinoma.3 Since 1973, CCC has been recognised in the WHO classification of ovarian tumours as a distinct histological entity, and its clinical behaviour is also distinctly different from that of other epithelial ovarian cancers.4 Several studies have shown that CCC patients exhibit a poor prognosis.5–9 In the literature, the low-level response of CCC to conventional taxane-based chemotherapy is associated with a poor prognosis.10 11 Several reports have shown that the lower-level proliferation of clear carcinoma cells may contribute to their resistance to chemotherapy.12 13 However, the mechanism of resistance to chemotherapy in CCC has remained unclear.

Glypicans are a family of heparan sulphate proteoglycans that are linked to the exocytoplasmic surface of the plasma membrane through a glycosylphosphatidylinositol anchor. Six glypicans have been identified in mammals (GPC1–GPC6), and two in Drosophila.14 The physiological function of glypicans is still not well understood. However, it was shown that the glypican-3 (GPC3)-encoding gene is mutated in patients with Simpson–Golabi–Behmel syndrome, an X-linked disorder characterised by prenatal and postnatal overgrowth and a varying range of dysmorphisms. GPC3 regulates cell growth either positively or negatively depending on the cell type. Genetic and functional studies showed that glypicans regulate the signalling activity of various morphogens, including Wnts, hedgehogs, bone morphogenic proteins and fibroblast growth factors.15–19 Previous studies showed that GPC3 was overexpressed in Wilms' tumour, hepatocellular carcinoma and hepatoblastoma.20 21 In ovarian carcinoma, GPC3 was overexpressed in yolk sac tumour and CCC, and not in other histotypes of EOC.22–24 In a previous report, we demonstrated that GPC3 was associated with taxol resistance in CCC.25 Thus, we hypothesised that GPC3 expression was associated with a poor prognosis in CCC patients. Maeda et al reported that GPC3 expression was significantly associated with a poor prognosis in stage III/IV CCC cases. However, the majority of clear cell adenocarcinomas are diagnosed at an early stage, and the relationship between GPC3 expression and the prognosis has remained unclear in early-stage CCC cases.

In the present study, we examined the immunohistochemical expression of GPC3 in CCC tissues to determine whether GPC3 expression is correlated with clinicopathological factors or the prognosis of CCC patients, especially in early-stage disease. We also investigated whether GPC3 was associated with CCC proliferation based on immunohistochemistry.

Materials and methods

Patients and tissue samples

Fifty-two human CCC tissue samples were obtained from patients who had undergone surgical treatment at Nagoya University Hospital between 1992 and 2006 after giving informed consent. The age of the patients ranged from 27 to 77 years, with a median of 52 years. None of these patients had undergone neoadjuvant chemotherapy before surgery.

All tissue samples were fixed in 10% formalin, embedded in paraffin and routinely stained with H&E for histological examination. All patients received postoperative chemotherapy with platinum plus cyclophosphamide and doxorubicin (before 1997) or platinum plus paclitaxel (after 1997). Tumour recurrence/progression was defined based on the clinical, radiological or histological diagnosis. Chemoresistance was defined as the appearance of a new lesion or a greater than 25% increase in tumour size within 6 months after finishing chemotherapy.


Formalin-fixed, paraffin-embedded tissue sections were cut at a thickness of 4 μm. For heat-induced epitope retrieval, deparaffinised sections in 0.01 M citrate buffer (Target Retrieval Solution, pH 6.1, Dako, Glostrup, Denmark) were treated three times at 90°C for 5 min using a microwave oven. Immunohistochemical staining was performed using the avidin-biotin immunoperoxidase technique (Histofine SAB-PO kit, Nichirei, Tokyo, Japan). Endogenous peroxidase activity was blocked by incubation with 0.3% hydrogen peroxide in methanol for 15 min, and non-specific immunoglobulin binding was blocked by incubation with 10% normal goat serum for 10 min. The sections were incubated at 4°C for 12 h with primary antibody against human GPC3 (1:200, clone IG12; BioMosaics, Burlington, Vermont, USA) and Ki-67 (1:200, clone MIB-1, Dako). The sections were rinsed and incubated for 30 min with biotinylated secondary antibody. After washing, the sections were incubated for 30 min with horseradish peroxidase-conjugated streptavidin and finally treated with 3-amino-9-ethylcarbazole in 0.01% hydrogen peroxide for 10 min. The slides were counterstained with Meyer's haematoxylin. The immunostaining intensity of GPC3 was scored semiquantitatively based on the per cent positivity of stained cells employing a 4-tiered scale as follows: for the evaluation of GPC3 expression, the staining intensity was scored as 0 (negative), 1 (weak), 2 (medium) or 3 (strong). The extent of staining was scored as 0 (0%), 1 (1–10%), 2 (11–50%) or 3 (≥51%) according to the percentage of the positive staining areas in relation to the total cancer areas. The sum of the intensity and extent scores was used as the final staining score (0–6) for GPC3. Tumours with a final staining score of more than 3 were considered to show positive expression. The scoring procedure was carried out twice by two independent observers (each blinded to the other's score) without any knowledge of the clinical parameters or other prognostic factors. The concordance rate was over 95% between the observers.

Statistical analysis

The χ2 test was also used to analyse the distribution of GPC3-positive cases, according to clinicopathological parameters. Survival analyses were conducted according to the life tables and Kaplan–Meier methods. Comparison of the survival between groups was performed with the log-rank test. Stat View software V.5.0 (SAS Institution, Cary, North Carolina, USA) was used for all statistical analyses, and p<0.05 was considered significant.


Immunohistochemical expression of GPC3 in EOC tissues

As figure 1 shows, the immunoreactivity of GPC3 was detected at various levels. There was little immunoreactivity of GPC3 in the tumour stroma. Among the 52 CCC specimens examined in this study, GPC3 was detected in 21 cases (40.4%) (table 1). GPC3 immunoreactivity, when categorised into negative versus positive expression, was not associated with the age, FIGO stage CA125 value, ascitic fluid volume or peritoneal cytology among the clinicopathological parameters tested (table 2). There were significant correlations between GPC3 expression and the residual tumour size. Residual tumour sizes were all within 1 cm in GPC3-positive cases; in seven of 24 patients, the residual tumour size was more than 2 cm in GPC3-negative cases.

Figure 1

Immunohistochemical staining patterns for glypican-3 (GPC3) in clear cell carcinoma. (A) Strong positive expression of GPC3. (B) Moderate positive expression of GPC3. (C) Weak positive expression of GPC3. (D) Negative expression of GPC3. (E) Positive control for GPC3 (normal placenta).

Table 1

Immunohistochemical staining in clear cell carcinoma patients

Table 2

Relationship between the expression of glypican-3 (GPC3) and clinicopathological parameters of clear cell carcinoma

Figure 2

Progression-free survival (PFS) and overall survival (OS) curves drawn using the Kaplan–Meier method according to glypican-3 (GPC3) expression in stage I clear cell carcinoma patients. (A) PFS. (B) OS. Borderline significant differences in PFS (p=0.05).

Correlation of GPC3 expression with survival of CCC patients

The median overall survival (OS) of CCC patients was 54.4 months (range 4.3–199.4). In univariate analyses, FIGO stage, residual tumour presence after primary cytoreductive surgery, positive peritoneal cytology and ascitic fluid volume were significant predictors of both a poor OS and poor progression-free survival (PFS); the CA125 value was the only significant OS predictor (table 3). The 5-year OS rates of patients with a negative (n=31) and positive (n=21) expression of GPC3 were 61.1% and 66.5%, respectively (table 3). No significant association was observed between GPC3 and survival (p=0.52). However, in stage I cases, the PFS of GPC3-positive was poorer than that of GPC3-negative CCC patients, although this was not significant (p=0.05; figure 2A), and the OS of GPC3-positive tended to be poorer than that of GPC3-negative CCC patients (p=0.11; figure 2B).

Table 3

Univariate analyses of several clinicopathological parameters in relation to the survival of patients with clear cell carcinoma

Figure 3

Immunohistochemical staining patterns for glypican-3 (GPC3) and Ki-67 in clear cell carcinoma; staining pattern of a tumour. (A) GPC3-positive. (B) GPC3-negative. (C, D) Ki-67. H&E staining was performed simultaneously for A and B. (E, F).

Correlation of GPC3 expression with cell proliferation

On immunohistochemical analysis of Ki-67 in GPC3-expressing patients, Ki-67-stained cells in GPC3-positive areas were markedly less frequent than those in GPC3-negative areas (figure 3). This suggested that GPC3 was associated with low-level proliferation in CCC of the ovary.

Clinical features in taxane-resistant patients

Thirty-six of 52 CCC patients received taxane-based chemotherapy. Twelve patients were resistant to this chemotherapy. Table 4 summarises the clinical features of the 12 taxane-based chemotherapy-resistant patients. Three of them (27.3%) were positive for GPC3 expression. The FIGO stages of GPC3-expressing patients were all stage I. In stage I patients, 19 received taxane-based chemotherapy. There were ten GPC3-positive and nine GPC3-negative patients. The chemoresistance rate was 30% in GPC3-positive and 11% in GPC3-negative patients. This suggested that GPC3 expression tended to be associated with resistance to taxane-based chemotherapy in early-stage CCC patients.

Table 4

Clinical features of taxane-resistant patients


A previous study showed that GPC3 is exclusively overexpressed in CCC among ovarian adenocarcinomas. In this study, GPC3 was expressed in 40.4% of CCC. Maeda et al showed that GPC3 expression was significantly associated with poor overall survival in stage III/IV CCC patients. However, it remained unclear whether GPC3 was associated with prognosis in early-stage CCC patients. Thus, we evaluated whether GPC3 expression was correlated with prognosis in CCC patients. In this study, clinicopathological analysis of CCC indicated that GPC3 expression was associated with PFS in patients with stage I (p=0.05), although the number of stage I patients was small. However, several limitations of this investigation should be noted.

First and foremost are the limitations inherent to the reliability and reproducibility of immunohistochemical techniques. Immunohistochemistry is semiquantitative and highly dependent on a range of poorly controlled variables, including the antibody concentration, choice of antibody, variability in the interpretation and stratification criteria, and inconsistency in specimen handling and technical procedures. Another limitation of immunohistochemical staining is the variability in the commonly used visual scoring system. These scoring methods are subjective, and so subject to human variability.

In general, the prognosis of CCC patients is poor, largely due to a low response rate to taxane-based chemotherapy. In this study, GPC3 expression tended to be associated with resistance to taxane-based chemotherapy in stage I patients. We have already reported that GPC3 is associated with resistance to paclitaxel.25 Our results showed the possible link between GPC3 expression and the paclitaxel-resistant nature of CCC in early-stage patients. Loss-of-function mutations of GPC3 are the cause of the X-linked Simpson–Golabi–Behmel syndrome (SGBS).26 This disorder is characterised by developmental overgrowth.27 GPC3-deficient mice also display developmental overgrowth along with several abnormalities found in SGBS patients.28 29 Because the cell sizes are similar in GPC3 null mice and their normal littermates, it has been concluded that the increase in body size in the absence of GPC3 is the result of a higher cell proliferation rate. Consequently, it is reasonable to propose that the developmental overgrowth of SGBS patients and GPC3-deficient mice indicates that GPC3 acts as an inhibitor of cell proliferation in the embryo. Likewise, GPC3 is an inhibitor of cell proliferation and can induce apoptosis in certain types of tumour cell. CCC showed a significantly lower proliferation compared with other histotypes of EOC. This finding may explain the high incidence of stage I patients with CCC. It is known that rapidly proliferating cells are the most sensitive, whereas cells that slowly proliferate are generally less sensitive to cytotoxic agents. CCC patients showed a very low-level response to chemotherapy and a high incidence of progressive disease. Accordingly, chemoresistance may be an important factor in the poor prognosis of CCC patients. Low proliferation activity may contribute to chemoresistance and the poor prognosis in CCC. This study suggested that GPC3 was associated with low-level proliferation, making CCC resistant to taxane-based chemotherapy. Further functional experiments, including an in vitro study, are needed to elucidate the relationship between GPC3 and proliferation.

In conclusion, we found that GPC3 expression was associated with a poor prognosis in early-stage CCC patients. These results indicate that GPC3 is a reliable and promising prognostic indicator in early-stage CCC patients, and might become a novel molecular target in the treatment strategy for CCC.

Take-home messages

  • In ovarian carcinoma, glypican-3 (GPC3) was overexpressed in clear cell carcinoma (CCC), and not in other histotypes of epithelial ovarian carcinoma.

  • GPC3 expression was associated with poor progression-free survival in stage I CCC patients.

  • Using immunohistochemistry, it was shown that GPC3 expression may be associated with a low proliferation rate in CCC cells.



  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.