Abstract
Background: The malignancy of gastrointestinal stromal tumours (GIST) varies greatly. Due to the uncertainty of specific molecular pathogenesis and complexity of biological behaviour of GIST, the aggressiveness and prognosis of GIST cannot be precisely predicted. Materials and Methods: We retrieved 40 paraffin-embedded specimens of surgically resected GIST between January 2013 and May 2015 at the Pathology Department of Yuhuangding Hospital Affiliated to Qingdao University and detected the expression of NIMA-interacting peptidylprolyl isomerase (PIN1) and Ki67, by immunohistochemical methods. Results: The positivity rate of PIN1 and Ki67 in GIST was 80% and 32.5%, respectively. The expression of PIN1 was associated with risk of malignancy, tumour location, tumour size, and mitotic counts. The expression of Ki67 was also associated with risk of malignancy, tumour location, tumour size, and mitotic counts. The expression of Ki67 was positively related to the expression of PIN1. Conclusion: PIN1 and Ki67 may be potential factors predicting the malignancy of GIST. PIN1 may be an attractive prognostic indicator and therapeutic target for GIST.
- Gastrointestinal stromal tumours
- NIMA-interacting peptidylprolyl isomerase
- Ki67 antigen
- immunohistochemistry
Gastrointestinal stromal tumour (GIST) is the most common mesenchymal tumour in the digestive tract, accounting for fewer than 1% of all gastrointestinal neoplasms. GIST can arise from any site of the gastrointestinal tract, with an estimated incidence of 10-20 million people annually worldwide. About 60% of GISTs occur in the stomach. Approximately 30% of GISTs arise in the small intestine, 10% in the oesophagus and rectum, and the remaining are extragastrointestinal. Their malignancy varies greatly; according to the modified National Institutes of Health (NIH) risk stratification it is classified into very low-risk, low-risk, intermediate-risk, and high-risk, based on tumour size, mitotic counts, primary tumour site and tumour rupture (1). Surgery remains the mainstay of treatment in treatment primary GIST. However, recurrences of the disease after surgery occur in approximately half of patients with GIST, including a minor proportion of patients with lower risk of malignancy. On the other hand, disease in some patients with higher risk did not recur, that could be due to specific molecular pathogenesis and complexity of the biological behaviour of GIST. This results in difficulties in predicting the aggressiveness and prognosis of the disease. Protein interacting with never in mitosis A1 (PIN1) is a human peptidyl-prolylcis-trans isomerase, interacting with never in mitosis A (NIMA) kinase. It specifically catalyzes the phosphorylated Ser/Thr-Pro motif to change its cis-trans isomerization, thereby modulating the conformation and function of its substrate proteins. Previous studies have verified that PIN1 is overexpressed in 38/60 tumour samples (2). Recent research demonstrated that overexpression of PIN1 contributes to centrosome amplification, chromosome instability and oncogenesis both in vivo and in vitro (3). PIN1 also participates in many cellular signaling pathways and promotes tumourigenesis as a catalyst or an amplifier.
Ki67 is a nuclear antigen closely involved in mitosis, which is essential for cell proliferation. PIN1 modulates many mitotic proteins and transcription factors, including Ki67. The hypothesis is that PIN1 not only participates in oncogenesis in GIST, but may also be a critical catalyst of tumour cell proliferation. This study aimed to investigate the immunochemical expression of PIN1 in GIST for the first time, and to explore its possible relationship with various clinicopathological features in patients with GIST.
Materials and Methods
Sample collection. Forty paraffin-embedded specimens of surgically resected GIST were retrieved between January 2013 to May 2015 from the Department of Pathology of Yantai Yuhuangding Hospital Affiliated to Qingdao University (Yantai, China). Definitive diagnoses were made on all patients, including 18 females and 22 males, with a median age of 58.5 years (range=40-83 years). Informed consent was obtained from all participants, and this study was approved by the Ethical Committee of the hospital. The diagnosis of the malignant potential of the tumour was according to the modified NIH risk classification depending on tumour size, mitotic count, primary tumour site and tumour rupture. All samples were divided into two groups, group I included patients with very low and low risk, group II those with intermediate and high risk. The slides for pathology examination were individually reviewed. Cell type was categorized as being predominantly spindle (>80% of the tumour), epithelioid (>80% of the tumour), or mixed if both the spindle and epithelioid accounted for >25% of all cancerous cells.
Primary antibodtes used in immunohistochemistry in this study.
Immnohistochemistry. Formalin-fixed and paraffin-embedded tissue specimens were sliced into 4-μm-thick sections for immunohistochemical analysis. Sections stained with haematoxylin and eosin were also reviewed. Each of the paraffin sections were de-waxed in xylene, and rehydrated through a series of ethanol gradients. The sections were then blocked with 3% hydrogen peroxide for 10 min, and subjected to antigen retrieval by microwave in 0.01 M citrate buffer for 30 min. The slides were blocked with 3% bovine serum albumin for 60 min, and then were washed three times with Tris-HCl Buffer Solution (TBS), and incubated with primary antibodies against CD117, Discovered on GIST-1 (DOG-1), CD34, smooth muscle actin (SMA), S-100, Ki67 and PIN1 (Table I) at 4°C overnight. After washing in TBS, the slides were incubated with second antibodies (poly peroxidase anti mouse/rabbit IgG). The antibody staining was visualized with 3,3-diaminobenzidine tetrahydrochloride (DAB; ZSGB-BIO, Beijing, P.R. China) in 0.1 M Tris (pH 7.2), containing 0.01% H2O2. The slides were counterstained with haematoxylin, dehydrated, and then mounted. The immunohistochemical reactions were then visualized under high-power magnification (×400) using an Olympus CH30 microscope (field width, 0.5 mm; Olympus Optical Co., Ltd., Tokyo, Japan). The primary antibody was replaced with TBS as negative control, and previous known positive slides of thyroid cancer also served as positive control.
Immunostaining scoring. Positive immunoreactivity was defined as brown yellow granules in the cytoplasm and/or nucleus of tumour cells under light microscopy. Under high power, more than 500 tumour cells in 5 fields (≥100 per field) for each slide were randomly examined.
All markers (CD117, DOG1, CD34, SMA and S-100) in the tumour cells on each slide were recorded as being expressed negatively or positively by estimating the number of positively stained tumour cells. Negative immunostaining was defined as fewer than 10% positively stained cells were detected, and positive immunostaining of Ki67 was defined when more than 10% of cells were positively stained.
Staining of PIN1 was defined as follows (4). Nuclear and cytoplasmic staining of PIN1 were classified into five grades (i.e. 0-4) and four grades (i.e. 0-3), respectively, by the visual intensity of PIN1 expression among all tumour cells. Cytoplasmic staining of PIN1 included: Grade 0, no staining; grade 1, weak staining/light yellow; grade 3, moderate staining/brown yellow; and grade 4, strong staining/dark brown. Nuclear staining of PIN1 was described as follows: Grade 0, positive cell rate=0; grade1,positive cell rate≤25%; grade 2, positive cell rate=26-50%; grade 3, positive cell rate=51-75%; and grade 4, positive cell rate >75%. The PIN1 immunoreactive score (IRS) was defined as the sum of both cytoplasmic and nuclear assessment, ranging from 0-7. Immunostaining of PIN1 was defined as positive when the score was greater than 3 .
Statistical analysis. SPSS 22.0 software (IBM, Armonk, NY, USA) was used to analyze the data. Correlation between expression of PIN1 and Ki67 was analyzed using the Spearman correlation test. The Mann–Whitney U-test or the Kruskal–Wallis was used to assess the association of the IRS for PIN1 and clinicopathological features. Association between various categorical variables and Ki67 positive rate was assessed by using Chi-square or Fisher's exact test. A p-value of less than 0.05 was considered statistically significant.
Results
Clinicopathological characteristics. Tumours were located in the stomach in 23 patients (57.5%), duodenum in one (2.5%), jejunum and ileum in 12 (30%), rectum in two (5%), abdominal cavity in one (2.5%), and in the liver in one (2.5%). The clinical manifestations at the time of diagnosis included abdominal pain in 17 patients (42.5%), gastrointestinal bleeding in nine (22.5%), abdominal discomfort in three (7.5%), and loss of appetite in one (2.5%). Nine patients were made aware of the tumour (22.5%) incidentally during ultrasonic, endoscopic, or physical examination, or other operations. According to the modified NIH risk stratification, six cases (15%) were classified as having very low risk, 12 (30%) low risk, 10 (25%) intermediate risk, and 12 (30%) high risk. Cell type consisted of spindle cells in 31 cases (77.5%), epithelioid cells in four cases (10%), and mixed-type in five cases (12.5%).
Immunoactive pattern of NIMA-interacting peptidylprolyl isomerase in gastrointestinal stromal tumour. A: Immunoreactive Score (IRS)=2. Cytoplasmic grade 0 and nuclear grade 2; B: IRS=3 cytoplasmic grade 1 and nuclear grade 2; C: IRS=4 cytoplasmic grade 1 and nuclear grade 3; D: IRS=5 cytoplasmic grade 2 and nuclear grade 3; E: IRS=6 cytoplasmic grade 2 and nuclear grade 4; F: IRS=7 cytoplasmic grade 3 and nuclear grade 4. Original magnification ×400.
Immunostaining analyses. The positivity of GIST were 100% to CD117, 100% to DOG-1, 85% to CD34, 17.7% to SMA, and 5% to S-100. On the whole, these markers were not clinicopathologically significant (data not shown). The positive rate of PIN1 expression was 80%. The positive signal was located in the nucleus and/or cytoplasm, and was diffusely distributed in most cases, although heterogeneously distributed in several cases (Figure 1). The IRS for PIN1 in group I was 4.22±1.22, whereas that of PIN1 in group II was 5.06±1.13; the difference between the two groups was statistically significant (p<0.05), indicating that PIN1 is more likely to be expressed in GIST of higher risk than of lower risk (Table II). Additionally, the IRS for PIN1 was significantly related to tumour location, mitotic counts, tumour size, and labelling index for Ki67 (p<0.05), but not associated with age, gender, or tumour necrosis (p>0.05) (Table II). The positive cells for Ki67 displayed brown-yellow granules only in the nucleus of tumour cells. The average positivity for Ki67 was 32.5% among all patients. In group II, Ki67 was positively expressed in 54.55% of the samples compared to 5.88% in group I (p<0.05). Moreover, the positivity for Ki67 was significantly associated with tumour size and mitotic counts (p<0.05), but was not related to age, gender, or tumour necrosis (p>0.05). Importantly, IRS for PIN1 was positively associated with positivity for Ki67 (p<0.05) (Table III).
Relationship between PIN1, Ki67 expression and clinicopathological characteristics.
Discussion
The biological behaviour and prognosis of GIST vary greatly, represented by differences in lesion size from less than 1 cm to over 10 cm in benign or malignant diseases. However, through long-term follow-up, almost any symptomatic GIST may behave aggressively (5). Based on a cohort study of 200 patients with GIST, DeMatteo et al. reported that about 46% of patients with complete resection of their tumour developed recurrence or metastasis within 5 years (6). Therefore, it is a huge challenge for clinicians and pathologists to precisely assess the malignancy of GIST.
Relationship between expression of NIMA-interacting peptidylprolyl isomerase (PIN1) and Ki67.
Previously, we evaluated the malignancy of GIST based on NIH risk stratification, including two parameters, tumour size and mitotic count (5). However, Joenuu et al. modified NIH risk stratification by adding extra two parameters to the system, primary tumour site and tumour rupture (1). For the past decade, researchers made a great effort to look for novel biomarkers for reliable evaluation of the biological behaviour of GIST, including cellularity and cell polymorphism, and also the expression of p21 (7), p16 (8), p53 (9), B-cell lymphoma-2 (BCL2) (10), epidermal growth factor receptor (EGFR), and cyclooxygenase2 (COX2) (11), amongst others. But only a limited number of markers has been discovered. Therefore, the discovery of new biomarkers is imperative to discriminate precisely those patients with GIST with malignant potential.
PIN1 specifically catalyses the phosphorylated Ser/Thr-Pro motif and affects the conformational and functional changes of target proteins, including protein activity and stability, phosphorylation status, dephosphorylation, protein–protein interactions, and subcellular localization (12). Overexpression of PIN1 was first found in breast cancer, and it was confirmed that PIN1 was closely related to malignancy of the tumour (13). A subsequent study indicated that compared to normal tissues, more than 10% cases showed overexpression in PIN1 in 38/60 types of human tumours, including prostate, lung, ovarian, cervical and brain cancer, and melanoma (2). Later research has demonstrated PIN1 to be overexpressed in many malignant tumours and to be closely associated with poor prognosis (14-18).
Little has been reported about PIN1 status in GIST. To our knowledge, this is the first study to investigate the relationship between PIN1 expression and the clinicopathological features immunohistochemically. The IRS for PIN1 in the high-risk group was significantly higher than that of the low-risk group (p<0.05). According to our results, PIN1 was expressed in all GISTs, and the intensity of PIN1 staining increased as the risk of GIST increased. Therefore, we assume that PIN1 may contribute to the oncogenesis and development of GIST, and its overexpression in higher-risk GIST may promote tumour cell growth and proliferation. Some studies have shown that tumour necrosis is related to the prognosis of GIST (19). However, this was not confirmed in this study as the correlation between PIN1 and tumour necrosis was not significant (p>0.05). The IRS for PIN1 was also significantly different by different tumour location (p<0.05), especially between the stomach and the small intestine (p<0.05).
It is reported that PIN1 influences various signaling pathways in cancer. PIN1 is significantly overexpressed in breast caner, and elevates cyclin D1 expression through activating its promoter. Additionally, PIN1 binds phosphorylated Ser63/73-Pro motifs on c-JUN by activated c-Jun N-terminal kinase (JNK) or RAS. Moreover, PIN1 collaborated with activated JNK or RAS to improve transcriptional activity of c-JUN towards the promoter of cyclin D1 promoter. All the above indicate that overexpression of PIN1 may promote growth of breast cancer by cooperating with RAS signaling and up-regulating cyclin D1 expression (13).
PIN1 also participates in many other oncogenic pathways, such as protein receptor tyrosine kinase RAS/extracelular signal-regulated kinase, and PI3/AKT pathway. Previous studies have indicated that tumour cells extracted from KIT-mutant GIST provided evidence of activation of downstream signaling pathways, including the mitogen-activated protein kinase pathway (RAF, MEK and ERK), the PI3/AKT pathway, and signal transducers and activators of transcription 3 (STAT3) (20), which PIN1 may influence in GIST.
PIN1 also activates many oncogenes and growth-promoting factors such as FOS and STAT3, which also play an important role in the transcription process of GIST, and inactivate many tumour suppressors and growth inhibitors. PIN1 may serve as catalyst and amplifier in the tumourigenesis of GIST, and work as on “on-off” switch for many oncogenes and suppressors simultaneously or in seperate processes, which may accelerate the progression of GIST.
Ki67 is a marker of cell proliferation, which is expressed in all stages of the cell cycle, except G0 stage. Ki67 is closely related to differentiation, infiltration, metastasis and prognosis in many tumours. There are studies indicating that the positive expression of Ki67 was intimately connected with the aggressive behaviour of GIST (21), and was a good predictive factor to evaluate the prognosis of GIST (22). However, the cut-off value of Ki67 to predict the outcome of GIST remains controversial. In accordance to the studies of Nakamura et al. (21) and Toquet et al. (23), the cut-off of Ki67 in this study was 10%, and demonstrates that the positive rate of Ki67 positively correlates to the malignant feature of the tumour, including mitotic count and tumour size (p<0.05). While other studies reported that when the cut-off was defined as 5%, the 1-, 3-, 5-year survival rates in patients with Ki67 positivity were statistically lower than in those with Ki67 negativity (24). Thus, the expression of Ki67 can be a predictor of unfavourable prognosis in GIST. Others drew the conclusion that a Ki67 index greater than 8% can precisely distinguish between high-risk GIST with different outcomes, after evaluating recurrence-free survival of 418 patients with GIST (25). Therefore, it was concluded that the Ki67 index can be an effective complement of modified NIH risk stratification in predicting prognosis of GIST.
In the cell cycle, PIN1 regulates many mitosis proteins, and also regulates many transcriptional factors and proteins, including Ki67 (26). Thus, PIN1 regulates the cell cycle through many mitosis-associated proteins and factors, and is closely correlated with cell proliferation. In this study, IRS of PIN1 in the Ki67-positive group was statistically higher than that of the Ki67-negative group (p<0.05). Further analysis indicated that PIN1 was positively related to Ki67 (r=0.371, p<0.05), which was in accordance with the literature mentioned above.
In conclusion, the expression of both PIN1 and Ki67 are closely associated with the prognosis of GIST, and they may both be potential indicators for the evaluation of the malignancy of GIST. The expression of PIN1 was positively associated with Ki67. PIN1 may be an attractive prognostic indicator and a therapeutic target for GIST. Further studies are warranted to verify the practical value of this potential biomarker for GIST and to reveal the specific mechanisms involved in growth and progression of this malignancy.
Footnotes
↵* These Authors contributed equally to this study Lixin Jiang,
Conflicts of Interest
No potential conflicts of interest were disclosed.
- Received January 4, 2016.
- Revision received February 17, 2016.
- Accepted February 18, 2016.
- Copyright© 2016 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.