Gastric cancer represents the second leading cause of cancer mortality worldwide. While surgical resection provides the only means of cure, the prevalence of advanced stage disease at presentation reflects the high mortality rate of this malignancy. But even after potentially curative resection, the five-year survival rates of 35–45% are still modest.1^–3 The positive impact of adjuvant chemoradiotherapy on survival in patients with resected carcinoma was reported in 2001 by Macdonald et al.4 However, the majority of patients develop tumour recurrence. Therefore, there is a need for new potential markers with therapeutic options to better tailor treatment. Cancer immunotherapy and molecular target therapy aimed at cancer specific genes describe a new way of treatment of different cancer entities.5^–7

The human trophoblast cell-surface antigen TROP2 (also termed GA733-1, M1S1, EGP-1) is encoded by the TACSTD2 gene which has been mapped to the human chromosome 1p32.8 9 The TACSTD gene family encodes for two different integral membrane glycoproteins, termed TACSTD2 (TROP2) and TACSTD1 (EpCAM). The TACSTD2 gene encodes a protein with several structural features in common with the transmembrane glycoprotein TACSTD1.10 TROP2, originally identified on human trophoblast and choriocarcinoma cell lines, was subsequently shown to be highly expressed by the majority of human malignant tumours.11 12 The function of TROP2 is not well understood although it has previously been suggested to play a role in regulating the growth of carcinoma cells.13 14 The cytoplasmic tail of TROP2 can be phosphorylated on serine 303 and contains a phosphoinositide binding sequence with a HIKE-like structure. Mutations in this sequence are found in different human genetic diseases and in cancer.15 We recently showed that TROP2 overexpression was a prognostic marker for poor survival in oral squamous cell carcinoma.16 As a consequence of its wide distribution on epithelial tumours, TROP2 seems to be an attractive target for immunotherapy. In the present study we examined the expression of TROP2 in the tissue of patients with resected gastric cancer and its correlation with clinicopathological features.

MATERIALS AND METHODS

Patients and tissue samples

The study was conducted according to the regulations of the local ethics committee. A total of 104 patients with gastric cancer resected between 1993 and 2004 at the Department of Visceral, Transplant and Thoracic Surgery, Medical University Innsbruck, were included in this retrospective study. Clinical data were obtained by reviewing the charts as well as from our computerised documentation system (ChiBASE). All data regarding date and cause of death were confirmed by the “Tumorregister Tirol”, a cancer register maintained by the Tyrolean government. Tumours were histologically classified according to the Lauren classification into intestinal and diffuse gastric adenocarcinomas and were staged according to the International Union Against Cancer classification (UICC). Adjuvant or additive chemotherapy was recommended to patients with UICC stage III and IV. For subgroup analysis, patients were divided into early stage disease (UICC IA+IB) and advanced stage disease (UICC stage >II). Table 2 summarises demographic data and tumour characteristics.

Immunohistochemistry

The expression of TROP2 was determined by immunohistochemistry using the purified goat polyclonal antibody against the recombinant human TROP2 extracellular domain at a dilution of 1:50 (AF650, R&D Systems, Minneapolis, Minnesota, USA). Briefly, 5 μm sections were cut from paraffin blocks, mounted on adhesive-coated glass slides (Instrumedics, Missouri, USA), deparaffinised and rehydrated with xylene and graded alcohol series. Endogenous peroxidase was blocked with methanol containing 30% hydrogen peroxide for 20 min. Pretreatment consisted of a 20 min incubation period in a pronase solution (water bath at 37°C). After several washes, sections were incubated in a biotinylated anti-goat antibody for 30 min (1:100, Ab6740, Abcam, Cambridge, UK). Labelling for TROP2 was completed using the ABC reagent (Vectastain Elite ABC Kit Standard PK-6100, Vienna, Austria) for 30 min at room temperature. Visualisation of immunolabelled structures was achieved by incubating sections in diaminobenzidine/H₂O₂ solution containing 30% hydrogen peroxide until coloration was visible (usually 3–4 min). Finally, slides were counterstained with Mayer’s hemalaun solution (Merck, Darmstadt, Germany), followed by clearing and mounting. Positive and negative controls were included in each run. Representative micrographs of tumours with predominant membranous staining of TROP2 are shown in fig 1.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1 Immunohistochemical staining for TROP2 expression in human gastric cancer tissues, original magnification ×100. Note the homogenous membranous immunostaining of tumour cells. (A) Gastric cancer without immunostaining. (B) Gastric cancer with strong (score 12) expression. (C) Normal gastric mucosa with immunostaining (arrow).

Evaluation of TROP-2 expression

Two independent pathologists (PM and HM) evaluated TROP2 expression in a blinded fashion using light microscopy. Discordant cases were re-evaluated to achieve a consensus. Antigen expression was defined as the presence of specific staining on the surface of tumour cells. TROP2 overexpression was evaluated by calculating a total immunostaining score as the product of a proportion and intensity score. The proportion score described the estimated fraction of positive stained tumour cells (0 = none; 1 = <10%; 2 = 10–50%; 3 = 51–80%; 4 = >80%). The intensity score represented the estimated staining intensity (0, no staining; 1, weak; 2, moderate; 3, strong). The total score ranged from 0 to 12. Regarding the total score the tissue samples were bimodally distributed with the nadir at a total score of 3–4. Therefore, TROP2 “overexpression” was arbitrarily defined as a total score >4 as described previously.17

Statistical methods

Statistical analysis was performed using SPSS V.10.0 (SPSS, Chicago, Illinois, USA). Survival curves were calculated using the Kaplan and Meier method and compared by the log-rank test. Follow-up time was censored if the patient was lost to follow-up. Patients who died without documented disease recurrence were considered censored for disease-free survival but were included as deaths for overall survival analysis. The association between TROP2 expression and clinicopathological variables were assessed by the χ² test. Factors shown to be of prognostic significance in the univariate models were evaluated in a multivariate Cox regression model. A p value <0.05 was considered statistically significant.

RESULTS

Surgical outcome

All patients underwent surgical intervention with curative intent. Total gastrectomy was performed in 47 patients whereas 57 patients had undergone subtotal gastrectomy. Gastric cancer with early stage disease (UICC IA+IB) was noted in 37 (36%) cases, while advanced cancer (UICC>II) was found in 67 (64%) cases. Chemotherapy was recommended to patients with UICC stage III and IV. In six patients neoadjuvant chemotherapy was used to facilitate radical resection. Among all patients, complete resection (R0) of the tumour was possible in 97 (93%) cases.

Histological subtypes

Intestinal-type gastric cancer was diagnosed in 88 (85%) cases whereas 16 (15%) patients had a diffuse carcinoma. Median overall survival time for patients with intestinal-type carcinoma was 52 months (range 1–163); it was 16 months (range 1–54) for diffuse-type carcinoma. For diffuse-type gastric cancer patients median disease-free survival was 14 (range 1–55) months (not reached in intestinal-type gastric cancer). Histological subtype significantly correlated with gender, age, grade, resection margins and overexpression of TROP2, whereas no correlation was found for depth of invasion, nodal stage and the presence of distant metastases. TROP2 overexpression could be detected in 60% of intestinal-type cancer and 31% of diffuse type (p = 0.03), respectively.

Immunohistochemistry

Normal gastric epithelial cells showed no to weak TROP2 staining. In contrast, in cancer cells a moderate to strong homogeneous membranous expression of TROP2 was noted in 58 (56%) carcinoma specimens (fig 1). Strong TROP2 expression (score 9–12) was found in 23 of 104 (22%) cases, whereas moderate expression (score 6–8) occurred in 35 (34%) cases and weak or no expression (score 0–4) in 46 (44%) cases. Thus, according to the above mentioned criteria, TROP2 was found to be overexpressed in 58 (56%) of 104 cases.

TROP2 expression and correlation with clinicopathological features

By univariate analysis we compared TROP2 overexpression to sex, histological grade, T stage, lymph node metastases, presence of distant metastases, resection margin and UICC stage (table 2).

In diffuse-type gastric cancer patients TROP2 overexpression correlated significantly with the depth of wall invasion of the tumour. In contrast, no correlation between TROP2 overexpression and clinicopathological parameters was found in carcinoma of intestinal-type.

Using the method of Kaplan and Meier, patients with intestinal-type carcinoma had a significantly better disease-free (p<0.01) and overall survival (p<0.01) than patients with diffuse-type gastric cancer. It should be noted that TROP2 overexpression significantly correlated with shorter disease-free survival (p = 0.03) in the subgroup of patients with intestinal-type gastric cancer.

Moreover, in patients with advanced intestinal-type gastric cancer (UICC ⩾II), TROP2 overexpression was significantly associated with poor disease-free (p<0.01) (fig 2A) and overall survival (p = 0.01) (fig 2B). In contrast, TROP2 overexpression was not predictive for disease-free and overall survival in patients with intestinal type gastric cancer in an early tumour stage (UICC IA+IB). Interestingly, among the total group, TROP2 overexpression was significantly associated with poor disease-free (p<0.01) (fig 2C) and overall survival (p = 0.03) (fig 2D) in patients with but not in patients without lymph node metastases.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2 (A) Prognostic significance of human trophoblast cell-surface antigen (TROP2) overexpression in 56 patients with advanced intestinal type cancer regarding disease-free survival by Kaplan–Meier analysis. high, patients with TROP2 overexpressing tumour (n = 32); low, patients with tumour (n = 24) without TROP2 overexpression. Patients with low TROP2 expression had a significantly better disease-free survival rate than patients with high TROP2 expression as defined by log-rank test (p<0.01). (B) Prognostic significance of human trophoblast cell-surface antigen (TROP2) overexpression in 56 patients with advanced intestinal type gastric cancer regarding overall survival by Kaplan–Meier analysis. high, patients with TROP2 overexpressing tumour (n = 32); low, patients with tumour (n = 24) without TROP2 overexpression. Patients with low TROP2 expression had a significantly better overall survival rate than patients with high TROP2 expression as defined by log-rank test (p = 0.01). (C) Prognostic significance of human trophoblast cell-surface antigen (TROP2) overexpression among all gastric cancer patients with lymph node metastasis (n = 58) regarding disease-free survival by Kaplan–Meier analysis. high, patients with TROP2 overexpressing tumour (n = 32); low, patients with tumour (n = 26) without TROP2 overexpression. Patients with low TROP2 expression had a significantly better disease-free survival rate than patients with high TROP2 expression as defined by log-rank test (p<0.01). (D) Prognostic significance of human trophoblast cell-surface antigen (TROP2) overexpression among all gastric cancer patients with lymph node metastasis (n = 58) regarding overall survival by Kaplan–Meier analysis. high, patients with TROP2 overexpressing tumour (n = 32); low, patients with tumour (n = 26) without TROP2 overexpression. Patients with low TROP2 expression had a significantly better overall survival rate than patients with high TROP2 expression as defined by log-rank test (p = 0.03).

Univariate survival calculations for the total cohort revealed depth (T stage) of wall invasion (p<0.01), nodal status (p<0.01), presence of distant metastasis (p<0.01), completeness of resection (R status) and UICC stage (p = 0.001) as significant predictors of disease-free survival, whereas gender (p = 0.27) and histological grade (p = 0.19) were not. There was a trend towards inferior disease-free survival in patients with TROP2 overexpression. Regarding overall survival, T stage (p<0.01), nodal status (p = 0.03), presence of distant metastasis (p<0.01), R status (p = 0.03), histological grade (p = 0.02) and UICC stage (p = 0.006)were of prognostic significance, whereas gender and TROP2 overexpression were not (table 3).

For multivariate analysis, all statistically significant variables evaluated in the univariate analyses were included in a Cox proportional hazard analysis. Remarkably, in intestinal type carcinoma TROP2 overexpression (p<0.01; relative risk 6.39; 95% CI 2.20 to 18.55) along with depth of wall invasion, lymph node metastases and involvement of free resection margin were independent prognostic factors for disease-free survival (table 4).

For patients with diffuse-type carcinomas multivariate analysis was omitted due to small sample size.

DISCUSSION

The present study showed the expression of TROP2 and its prognostic value in gastric cancer. For the first time, our data provide evidence that TROP2 plays an important role in this aggressive malignancy. Significantly higher expression of TROP2 could be detected in intestinal-type gastric cancer than in diffuse-type carcinoma (p = 0.03), although a comparison of the two groups is disputable because of the different number of cases. In intestinal-type gastric cancer TROP2 overexpression was detectable in 53 of 88 (60%) specimens and was an independent prognostic marker for disease-free survival. In univariate analyses TROP2 overexpression significantly correlated with poor disease-free and overall survival in advanced intestinal-type carcinoma. Interestingly, among all patients TROP2 overexpression significantly correlated with poor disease-free and overall survival in lymph node positive patients but not in lymph node negative patients.

TROP2, encoded by the single-exon gene TACSTD2 has been shown to play an important role in tumour growth. We recently reported that TROP2 overexpression predicts poor survival in oral squamous cell carcinoma.16 In addition, Nakashima et al showed that TROP2 is overexpressed in oesophageal cancer and suggested that its expression could be related closely to malignant transformation.18 The results of this study provide evidence that TROP2 overexpression predicts poor disease-free and overall survival, particularly in patients with intestinal type gastric cancer. Interestingly, a multistage model of carcinogenesis has been described for intestinal gastric carcinoma which is similar to the adenoma–carcinoma sequence reported for colon cancer.19 Our data are in line with results from Ohmachi et al who indicated that TROP2 was strongly expressed in colorectal cancer, correlating with aggressiveness and poor prognosis.20

Moreover we showed that TROP2 overexpression was predictive for disease recurrence and poor overall survival in lymph node positive gastric cancer patients and advanced intestinal type carcinoma.

Although TROP2 has been previously suggested to function as a cell–cell adhesion receptor in cancer cells, recent data are indicative of a function in cell signal transduction and tumour cell growth.21 Recently, Wang et al found evidence that TROP2 has a functional role in tumourigenesis. They showed that a series of colon carcinoma cell lines require TROP2 expression for tumourigenic growth. By knocking down TROP2 in certain colon cancer cell lines, the number of formed cell colonies was dramatically reduced compared to the control cell lines. Furthermore, suppression of TROP2 expression inhibited the invasive behaviour of a certain cell line. Conversely, increasing TROP2 expression led to increased anchorage-independent growth of colon cancer cell lines.22 TROP2 is highly homologous to the cell–cell adhesion molecule Ep-CAM.15 Due to immunogenic properties of Ep-CAM in cancer patients, active and passive immunotherapeutic agents have been developed.23 In resected colon cancer patients, significant reduction in mortality and relapse rates can be induced following treatment with the Ep-CAM-specific monoclonal antibody edrecolomab.24 The immunogenic properties have been recently described by Mangino et al who showed TROP2 to be a target molecule efficiently recognised by human cytotoxic T lymphocytes.25 Furthermore, a humanised monoclonal antibody (hRS7) was generated with promising results in an in vivo breast cancer model.26

In summary, we have found evidence that TROP2 is associated with gastric cancer growth. Due to its wide distribution TROP2 may become an attractive therapeutic target in a subgroup of patients with gastric cancer. Moreover, based on the predictive value for recurrence, TROP2 may also practically be used as an immunohistochemical marker for a more aggressive treatment strategy.