The Clinical Implication of PTEN and FAK Expression in Gastric Cancer Patients

Despite the declining incidence of gastric cancer (GC), this disease remains the sixth most common cancer and the second most common cause of cancer-related death.¹ Surgical resection and lymph node dissection are the main treatments for curing GC, and the pathologic tumor stage was associated with patient prognosis.

Tumor suppressor phosphatase and tensin homolog (PTEN) expression was reported to be associated with good prognosis in GC.² The clinical implications of focal adhesion kinase (FAK) expression on cancer patient prognosis are controversial. Overexpression of FAK was associated with a poor prognosis in colon cancer,³ pancreatic cancer,⁴ breast cancer, and GC,⁵ whereas low FAK expression was associated with a poor prognosis in cholangiocarcinoma.⁶

Overexpression of PTEN has been reported to lead to the downregulation of FAK expression and inhibited GC cell invasion.⁷ Furthermore, downregulation of Notch1 could inhibit the invasion and metastasis of GC through PTEN activation and dephosphorylation of AKT (a serine/threonine protein kinase, also known as protein kinase B) and FAK.⁸ However, there are no reports regarding the correlations between PTEN and FAK expression and their impact on GC patient prognosis.

The aim of the present study was to analyze the correlation between PTEN and FAK expression, the clinicopathologic characteristics, initial recurrence pattern, and patient prognosis of GC after curative surgery.

Materials and Methods

A total of 200 GC patients receiving curative resection were enrolled. The present study was approved by the Institutional Review Board at our hospital (number: 2017-10-009AC) and in accordance with the Ethical Principles for Medical Research Involving Human Subjects, as outlined in The Declaration of Helsinki. The exclusion criteria included a history of gastric surgery or a pathologic diagnosis other than adenocarcinoma. This study was approved by the Institutional Review Board of Taipei Veterans General Hospital and in accordance with the Ethical Principles for Medical Research Involving Human Subjects, as outlined in The Declaration of Helsinki.

The pathologic staging of GC was performed after surgery according to the 8th American Joint Committee on Cancer (AJCC)/Union for International Cancer Control (UICC) tumor, node, metastasis (TNM) classification.⁹ The data were prospectively collected and regularly updated. Before surgery, chest radiography, abdominal sonography, or abdominal computed tomography (CT) scan were performed. A total or subtotal gastrectomy was performed according to the tumor location.

After surgery, follow-up studies including image studies and tumor markers were arranged every 3 months for the first 3 years, followed by a follow-up every 6 months until the patient's death. In our hospital, adjuvant chemotherapy was not routinely performed before 2008. Since 2008, adjuvant chemotherapy such as TS-1 was performed for stage II and III GC patients. However, none of the patients enrolled in this study received adjuvant chemotherapy.

Recurrence was classified as locoregional, hematogenous, distant lymphatic, or peritoneal. Patients who experienced recurrence of GC after surgery could receive 5-fluouracil–based chemotherapy.

Immunohistochemical staining for PTEN and FAK protein

The procedures of immunohistochemical (IHC) staining were the same as a previous study.¹⁰ The monoclonal antibodies against PTEN (#9188; Cell Signaling Technology, Danvers, MA, USA) and FAK (#3285; Cell Signaling Technology) were used.^11,12 The slides were incubated with the antibody at a final dilution of 1:150 in phosphate-buffered saline overnight at 4°C. The reaction was visualized with AEC(RED) substrate Kit (cat. 00-2007, Invitrogen). Subsequently, the slides were briefly counterstained with hematoxylin. PTEN expression was evaluated in normal mucosa and adenocarcinomas. PTEN expression was observed in the cytoplasm of GC tissue. As in a previous report,¹⁰ the results of IHC staining for PTEN were scored as 0 (<5% positive cells), 1+ (5–25% positive cells), 2+ (26–75% positive cells), and 3+ (>75% positive cells). Low PTEN expression was defined when the IHC staining result was 0 or 1+, and high PTEN expression was defined when the IHC staining result was 2+ or 3+. The same criteria were applied in the definition of FAK expression. High and low expressions of PTEN and FAK are shown in Fig. 1.

View Figure

• Download as Powerpoint
• Download Figure

Results

Clinicopathologic features

Among the 200 GC patients, 144 exhibited high PTEN expression, and 56 had low PTEN expression. Patients with high PTEN expression were more likely to have smaller tumor size, more well- and moderately differentiated tumors, a more superficial gross appearance, less scirrhous stromal reactions, more likely to have high FAK expression, and have less advanced pathologic T category, N category, and TNM stage than patients with low PTEN expression (Table 1).

Table 1 Clinical profile in 200 gastric cancer patients with high or low PTEN expression

View Fullscreen

Among the 200 GC patients, 175 had high FAK expression, and 25 had low FAK expression. However, there was no significant difference in the clinicopathologic characteristics between patients with high FAK expression and low FAK expression (Supplemental Table 1).

We then divided GC patients into 4 groups according to PTEN and FAK expression: PTEN(low)/FAK(low), PTEN(low)/FAK(high), PTEN(high)/FAK(low), and PTEN(high)/FAK(high). Compared with GC patients with PTEN(low)/FAK(low) and PTEN(low)/FAK(high), GC with PTEN(high)/FAK(low) and PTEN(high)/FAK(high) had more well and moderate differentiation, a more superficial gross appearance, less scirrhous reactivity, and less advanced pathologic N category and TNM stage (Table 2).

Table 2 Clinical profile in 200 gastric cancer patients according to PTEN/FAK expression

View Fullscreen

Initial recurrence pattern

Patients with high PTEN expression had less tumor recurrence than those with low PTEN expression (25.0% versus 50.0%, P = 0.001). With regard to the initial recurrence pattern, patients with low PTEN expression were more likely to have anastomosis recurrence and distant metastasis (especially peritoneal recurrence and bone metastasis) than those with high PTEN expression (Supplemental Table 2).

Patients with high FAK expression had more tumor recurrence (52.8% versus 22.2%, P = 0.021) and distant metastasis (39.8% versus 11.1%, P = 0.018) than those with low FAK expression, especially for stage II and III GC. There was no significant difference in the initial recurrence pattern between patients with high FAK expression and patients with low FAK expression in stage I GC.

Patients with PTEN(high)/FAK(low) had the lowest tumor recurrence rate, followed by those with PTEN(high)/FAK(low), PTEN(low)/FAK(low), and PTEN(low)/FAK(high). Patients with PTEN(low)/FAK(high) had the most distant metastasis compared with that of the other three groups (P = 0.010). Patients with PTEN(high)/FAK(high) had no bone metastasis, which was the least compared with that of the other three groups (Table 3).

Table 3 Patterns of initial recurrence of gastric cancer after curative surgery according to PTEN/FAK expression

View Fullscreen

OS

As shown in Fig. 2a, the 5-year OS (60.2% versus 31.2%, P < 0.001) was significantly better in patients with high PTEN expression than low PTEN expression.

View Figure

• Download as Powerpoint
• Download Figure

As shown in Fig. 2c, the 5-year OS was similar between patients with high FAK expression and low FAK expression (50.4% versus 64.0%, P = 0.150). Regarding TNM stage, the 5-year OS was similar between patients with high FAK expression and low FAK expression in stage I GC (77.2% versus 71.4%, P = 0.692); however, the 5-year OS was higher in patients with low FAK expression than high FAK expression in stage II and III GC (61.1% versus 33.9%, P = 0.020).

In Fig. 3a, the 5-year OS was the highest in GC patients with PTEN(high)/FAK(low), followed by those with PTEN(high)/FAK(high), PTEN(low)/FAK(low), and PTEN(low)/FAK(high) (83.3% versus 58.0% versus 46.2% versus 26.5%, respectively, P < 0.001).

View Figure

• Download as Powerpoint
• Download Figure

Multivariate analysis demonstrated that age, sex, lymphovascular invasion, pathologic TNM stage, and PTEN/FAK expression status were independent prognostic factors for OS (Table 4).

Table 4 Multivariate Cox proportional-hazards model for the survival analysis of the gastric cancer patients after curative surgery

View Fullscreen

Discussion

This study is the first to investigate the correlation between PTEN and FAK expression and the clinicopathologic characteristics of GC patients. Our results showed that high PTEN expression is associated with a favorable patient prognosis and is an independent prognostic factor that affects OS and DFS. Patients with PTEN(high)/FAK(low) tumors were associated with better OS and DFS, followed by those with PTEN(high)/FAK(low), PTEN(low)/FAK(low), and PTEN(low)/FAK(high) tumors.

It was reported that downregulation of PTEN was correlated with peritoneal recurrence in GC,^13,14 which was associated with the activation of the PI3K/AKT pathway.¹³ In renal cell carcinoma,¹⁵ patients with bone metastasis were reported to have lower PTEN expression than those without. Bone metastasis might be caused by AKT and integrin α5 signaling. Our results showed that PTEN expression was associated with more anastomosis recurrence, peritoneal metastasis, and bone metastasis, which is similar to the above reports. Furthermore, a higher incidence of anastomosis recurrence in patients with low PTEN expression might be due to patients with low PTEN expression presenting a more advanced T category. Although this study showed significantly more bone metastasis in patients without PTEN expression than those with, the number of patients with bone metastasis was limited. The conclusion might not be reliable, although the statistical value is considerable. However, few studies^13,14 investigated the correlation between PTEN expression and recurrence pattern in GC, and the patient number enrolled is limited; this study enrolled the largest population to date to investigate their relationship. The enrollment of more patients is required to verify our findings in the future.

Our data showed that, in comparison with GC patients with PTEN(low)/FAK(low) and PTEN(low)/FAK(high), GC with PTEN(high)/FAK(low) and PTEN(high)/FAK(high) had more well and moderate differentiation, a more superficial gross appearance, less scirrhous reactivity, and less advanced pathologic N category and TNM stage. In Supplemental Table 1, there is no significant difference in the clinicopathologic characteristics between patients with high and low expression of FAK. It seems that low expression of PTEN was associated with more favorable pathologic characteristics, regardless of the expression of FAK.

It was reported that overexpression of PTEN could lead to downregulation of FAK and inhibit GC cell invasion.⁷ In this study, patients with high PTEN expression were more likely to have high FAK expression; however, whether PTEN or FAK expression would impact each other regarding patient prognosis is still unknown. Interestingly, both OS and DFS demonstrated that GC with PTEN(high)/FAK(low) were associated with the best prognosis, followed by GC with PTEN(high)/FAK(high), PTEN(low)/FAK(low), and PTEN(low)/FAK(high). Regarding tumor recurrence, GC with PTEN(high)/FAK(low) had the lowest tumor recurrence rate, followed by those with PTEN(high)/FAK(high), PTEN(low)/FAK(low), and PTEN(low)/FAK(high). In the present study, stage II and stage III GC patients with high FAK expression were associated with more tumor recurrence and a worse survival than patients with low FAK expression. It seems that PTEN expression was associated with a favorable prognosis and that FAK expression has an adverse effect on patient survival in GC. Further in vivo and in vitro studies regarding this issue are required to validate our results.

It was reported that miRNA-575 can target PTEN and regulate the development of GC.¹⁶ miRNA-1224 can inhibit tumor metastasis by targeting FAK in intestinal-type GC.¹⁷ Furthermore, miRNA-147 can inhibit cell proliferation and increase the chemosensitivity of GC to 5-FU by targeting PTEN.¹⁸ It seems that the status of PTEN and FAK expression may be helpful for evaluating the response to target therapy and chemotherapy. Our study demonstrated distinguishable patient prognoses according to the PTEN/FAK expression, which may provide useful information and have clinical impact on GC treatment.

There are some limitations in the present study. First, as it is a retrospective study, some bias is likely to exist. Second, although our results showed that tumors with low PTEN expression had significantly more bone metastasis than tumors with high PTEN expression, the number of patients with bone metastasis is limited, and further study is required for validation. Third, GC is heterogeneous, and immunohistochemistry might have false-positive and false-negative staining.

Conclusions

Our results demonstrated that GC patients with low PTEN expression have more distant metastasis and a worse prognosis than those with high PTEN expression. PTEN and FAK might have opposing effects on survival in GC. For high-risk GC patients, the physicians should be aware of the possibility of tumor recurrence. Our results may have clinical impact on treatment and follow-up for GC patients.