GLI1 and PTTG1 expression in colorectal carcinoma patients undergoing radical surgery and their correlation with lymph node metastasis

Abstract

BACKGROUND

Few studies have investigated the expression of GLI1 and PTTG1 in patients undergoing radical surgery for colorectal carcinoma (CRC) and their association with lymph node metastasis (LNM). Therefore, more relevant studies and analyses need to be conducted.

AIM

To explore GLI1 and PTTG1 expression in patients undergoing radical surgery for CRC and their correlation with LNM.

METHODS

This study selected 103 patients with CRC admitted to our hospital between April 2020 and April 2023. Sample specimens of CRC and adjacent tissues were collected to determine the positive rates and expression levels of GLI1 and PTTG1. The correlation of the two genes with patients’ clinicopathological data (e.g., LNM) was explored, and differences in GLI1 and PTTG1 expression between patients with LNM and those without were analyzed. Receiver operating characteristic (ROC) curves were plotted to evaluate the predictive potential of the two genes for LNM in patients with CRC.

RESULTS

Significantly higher positive rates and expression levels of GLI1 and PTTG1 were observed in CRC tissue samples compared with adjacent tissues. GLI1 and PTTG1 were strongly linked to LNM in patients undergoing radical surgery for CRC, with higher GLI1 and PTTG1 levels found in patients with LNM than in those without. The areas under the ROC curve of GLI1 and PTTG1 in assessing LNM in patients with CRC were 0.824 and 0.811, respectively.

CONCLUSION

GLI1 and PTTG1 expression was upregulated in patients undergoing radical surgery for CRC and are significantly related to LNM in these patients. Moreover, high GLI1 and PTTG1 expression can indicate LNM in patients with CRC undergoing radical surgery. The expression of both genes has certain diagnostic and therapeutic significance.

Key Words: Colorectal carcinoma, GLI1, PTTG1, Radical resection, Lymph node metastasis

Core Tip: The main goal of radical surgery for colorectal carcinoma (CRC) is to completely remove the primary tumor and regional lymph nodes. Lymph nodes are essential in predicting patient prognosis and for determining the need for adjuvant chemotherapy after surgical resection. In this study, the expression of GLI1 and PTTG1 in patients with CRC undergoing radical surgery and their correlation with lymph node metastasis (LNM) were examined. Aberrant overexpression of GLI1 and PTTG1 was confirmed in patients undergoing radical surgery for CRC. In addition, both GLI1 and PTTG1 were significantly related to LNM and could be used as predictors of LNM, indicating their clinical significance in the evaluation of CRC cases and the development of treatment plans.

INTRODUCTION

Colorectal carcinoma (CRC), which originates from the colon or rectum, is the third most common malignancy in both men and women worldwide[1]. Overweight, family history, tall stature, excessive intake of alcoholic beverages, and lack of regular exercise are the predisposing factors[2]. Etiologically, CRC is associated with excessive mucosal epithelial cell proliferation, distant tumor metastasis, long-term intestinal inflammatory responses, and abnormal changes in the gut microbiota[3,4]. The global number of new CRC cases in 2018 was 1.8 million, with nearly 900000 deaths[5]. Patients with CRC may present with clinical symptoms such as celialgia, abdominal mass, hemoproctia, abnormal weight loss, abnormal appetite loss, depression, and insomnia, which can seriously affect their physical and mental health and their quality of life[6,7]. Radical surgery for CRC is a common treatment option that aims for a complete resection of the primary tumor and regional lymph nodes. Lymph nodes are essential for predicting patient outcomes and determining the need for adjuvant chemotherapy after surgical resection[8].

GLI1 is a gene that mediates pathophysiological processes and changes in various tumors. It is involved in the proliferation, migration, invasiveness, growth, vascular invasion, and self-renewal of cancer stem cells in various malignancies[9,10]. According to Du et al[11], GLI1, as a component of the hedgehog (Hh)/GLI1 signaling pathway in CRC, can affect the inhibition of resveratrol on the growth and metastasis of HCT116 cells. Park et al[12] found that GLI1 influences the survival of patients with CRC by participating in hypoxic mechanisms and interacting with CCT2. PTTG1, a carcinogenic agent closely associated with tumor metastasis, participates in organ development, metabolism, DNA damage/repair, and cell cycle regulation. It is aberrantly overexpressed in various tumors, including those in the pituitary gland, thyroid, breast, ovary, uterus, and colon[13,14]. The promotion by PTTG1 of tumor metastasis and invasion has been linked to its direct regulation of basic fibroblast growth factor, epidermal growth factor, and β-catenin/transcription factor[15].

Research on the expression of GLI1 and PTTG1 in patients with CRC undergoing radical surgery and their correlation with lymph node metastasis (LNM) is limited. This study sought to bridge this gap to contribute to the improved clinical management of CRC.

MATERIALS AND METHODS

General information

We enrolled 103 patients who underwent radical surgery for CRC at Shanghai Zhangjiang Institute of Medical Innovation between April 2020 and April 2023. Samples of CRC and adjacent tissues were collected from all patients. Of the 103 patients, 53 were male and 50 were female. Furthermore, 42 were < 60 years old, and 61 were ≥ 60 years old. Tumors were located in the colon in 43 cases and in the rectum in 60 cases. In 23 cases, tumors were well differentiated, whereas in 80 cases, they were moderately and poorly differentiated. Furthermore, 48 cases were pathologically staged I–II, whereas 55 were staged III–IV. LNM was observed in 58 cases.

Criteria for patient enrollment and exclusion

The inclusion criteria were as follows: Diagnosis of CRC by surgical pathology and colonoscopic biopsy; met corresponding indications for surgical treatment and underwent radical surgery, with complete cancer and para-cancer tissue specimens obtained; no mental and cognitive abnormalities; and intact clinical data.

The exclusion criteria were as follows: systemic inflammatory diseases; previous use of chemoradiotherapy or targeted therapy; presence of other malignant tumors; estimated survival < 3 months; inability to attend follow-up visits or loss to follow-up; lactating or pregnant women; infectious diseases or blood disorders.

Sample collection and testing methods

Samples of CRC and adjacent tissues of about 2–3 g were collected from all patients during the surgery and subsequently processed into sections (5–10 μm thick) after preliminary treatment. The specimen sections were placed in an electrothermal constant-temperature air-blowing drying oven at 50°C–60°C for 3–4 h to make them adhere to the slide. After baking, disodium hydrogen phosphate–sodium dihydrogen phosphate buffer was used for antigen retrieval, and mouse antihuman GLI1 and PTTG1 monoclonal antibodies were added overnight. Next, each tissue sample was stained by immunohistochemistry following the SP kit instructions. Then, the positive rates of GLI1 and PTTG1 expression in each tissue sample were observed using an automatic immunohistochemical staining instrument.

The criteria for judging the positive expression rate were as follows: (1) According to the product of the staining intensity and the number of positive cells in the field of view, a percentage of positive cells < 5% was counted as 0 points, 5%–25% as 1 point, 26%–50% as 2 points, 51%–75% as 3 points, and > 75% as 4 points; and (2) Based on dyeing intensity, 0, 1, 2, and 3 points were given for colorless, light yellow, brownish yellow, and tan, respectively. Positive expression was determined if the sum of the two results was greater than 3 points. The immunohistochemical staining instrument was set to counting mode in counting the number of GLI1- and PTTG1-positive cells in the visual field. Finally, the expression levels of GLI1 and PTTG1 were calculated.

Outcome measures

The positive expression rates and protein levels of GLI1 and PTTG1 in CRC and adjacent tissues were statistically compared. The correlations between GLI1 and PTTG1 expression and clinicopathological characteristics, such as sex, age, tumor location, degree of tumor differentiation, pathological stage, and LNM, were evaluated. The differences in GLI1 and PTTG1 expression levels between patients with and without LNM were examined.

Statistical analysis

All analyses were performed using SPSS 20.0 at a significance level of P < 0.05. Categorical variables were expressed as number (percentage) and comparatively analyzed using the χ² test. Continuous variables were expressed as mean ± SE and analyzed using the independent sample t-test, respectively.

RESULTS

Positive rates of GLI1 and PTTG1 expression in CRC tissue samples

The positive expression rates of GLI1 in CRC and adjacent tissues were 63.11% and 12.62%, respectively, and the corresponding positive expression rates of PTTG1 were 56.31% and 5.83%, respectively. Significantly higher positive rates of GLI1 and PTTG1 were observed in cancer tissues than in adjacent tissues (P < 0.001; Table 1).

Table 1 Positive rates of GLI1 and PTTG1 in colorectal carcinoma tissue samples.

Grouping	n	Cancerous tissue	Para-cancerous tissue	χ2	P value
GLI1 positive	103	65 (63.11)	13 (12.62)	55.792	< 0.001
PTTG1 positive	103	58 (56.31)	6 (5.83)	61.292	< 0.001

GLI1 and PTTG1 expression levels in CRC tissue samples

Significantly higher GLI1 and PTTG1 levels were observed in cancer tissue samples than in adjacent tissues (P < 0.05; Figure 1).

Figure 1 Comparison of GLI1 and PTTG1 expression between colorectal carcinoma tissue samples. A: Para-cancerous tissue; B: Adjacent tissues. ^aP < 0.05.

Correlation of GLI1 and PTTG1 with patients’ clinicopathological data

Neither GLI1 or PTTG1 expression were strongly associated with sex, age, tumor location, degree of tumor differentiation, or pathological stage (P > 0.05). However, the expression of both genes was significantly associated with LNM (P < 0.05; Tables 2 and 3).

Table 2 Correlation of GLI1 expression with clinicopathological data of patients undergoing radical surgery for colorectal carcinoma.

Grouping	n	GLI1 positive (n = 65), n (%)	GLI1 negative (n = 38), n (%)	χ2	P value
Sex				0.403	0.526
Male	53	35 (53.85)	18 (47.37)
Female	50	30 (46.15)	20 (52.63)
Age				3.504	0.061
< 60 yr old	42	22 (33.85)	20 (53.63)
≥ 60 yr old	61	43 (66.15)	18 (47.37)
Tumor location				0.221	0.638
Colon	43	26 (40.00)	17 (44.74)
Rectum	60	39 (60.00)	21 (55.26)
Differentiation degree				0.552	0.458
High differentiation	23	13 (20.00)	10 (26.32)
Medium–low differentiation	80	52 (80.00)	28 (73.68)
Pathological staging				3.086	0.079
I–II	48	26 (40.00)	22 (57.89)
III–IV	55	39 (60.00)	16 (42.11)
Lymph node metastasis				5.233	0.022
Positive	58	44 (67.69)	17 (44.74)
Negative	45	21 (32.31)	21 (55.26)

Table 3 Association between PTTG1 expression and clinicopathological data of patients undergoing radical surgery for colorectal carcinoma.

Grouping	n	PTTG1 positive (n = 58), n (%)	PTTG1 negative (n = 45), n (%)	χ2	P value
Sex				2.728	0.099
Male	53	34 (58.62)	19 (42.22)
Female	50	24 (41.38)	26 (57.78)
Age				3.534	0.060
< 60 yr old	42	19 (32.76)	23 (51.11)
≥ 60 yr old	61	39 (67.24)	22 (48.89)
Tumor location				2.327	0.127
Colon	43	28 (48.28)	15 (33.33)
Rectum	60	30 (51.72)	30 (66.67)
Differentiation degree				0.867	0.352
High differentiation	23	11 (18.97)	12 (26.67)
Low–moderate differentiation	80	47 (81.03)	33 (73.33)
Pathological staging				0.616	0.433
I–II	48	29 (50.00)	19 (42.22)
III–IV	55	29 (50.00)	26 (57.78)
Lymph node metastasis				17.150	< 0.001
Positive	58	43 (74.14)	15 (33.33)
Negative	45	15 (25.86)	30 (66.67)

GLI1 and PTTG1 expression in patients with LNM

Patients with LNM showed obviously higher GLI1 and PTTG1 levels than those without. The area under the curve (AUC) of GLI1 for predicting LNM in patients with CRC undergoing radical surgery was 0.824 (sensitivity: 75.56%; specificity: 84.48%), and the optimal cut-off was 19.76. The AUC, sensitivity, specificity, and the optimal cut-off of PTTG1 for predicting LNM were 0.811, 68.89%, 82.76%, and 13.97, respectively (Figure 2 and Table 4).

Figure 2 Comparison of GLI1 and PTTG1 expression in patients with and without lymph node metastasis. A: GLI1 expression in patients with and without LNM; B: PTTG1 expression in patients with and without lymph node metastasis; C and D: Receiver operating characteristic. ^aP < 0.05.

Table 4 Receiver operating characteristic analysis of the potential of GLI1 and PTTG1 expression to indicate lymph node metastasis in patients undergoing radical surgery for colorectal carcinoma.

Grouping	AUC	95%CI	SE	Cut-off	Sensitivity (%)	Specificity (%)
GLI1	0.824	0.739–0.910	0.044	19.76	75.56	84.48
PTTG1	0.811	0.726–0.895	0.043	13.97	68.89	82.76

95%CI: 95% confidence interval.

DISCUSSION

Screening for CRC in patients based on early symptoms is difficult; imaging and pathological examination are often required[16]. However, in patients with CRC who underwent radical resection of the primary tumor, the accuracy of the said methods of disease assessment will be greatly affected[17]. To maximize the curative effect of radical surgery for CRC and promote patients’ postoperative recovery, more accurate and effective means of predicting disease changes are needed.

This study included 103 patients with CRC who underwent radical surgery and analyzed samples of cancer and adjacent tissues. First, we observed a significantly higher positive rate and expression of GLI1 in cancer tissue samples versus adjacent tissues, as well as a markedly higher positive rate and expression of PTTG1, suggesting the overexpression of GLI1 and PTTG1 and the possible involvement of both genes in tumorigenesis in CRC. Wu et al[18] found high GLI1 expression in human small-cell lung cancer, which indicates that GLI1 may mediate the onset and progression of small-cell lung cancer by activating the target gene downstream of Hh signaling. Chen et al[19] also reported high GLI1 expression in hepatocellular carcinoma cells and found that downregulating its expression inhibited the adhesion, movement, migration, and invasion of hepatocellular carcinoma cells. Ding et al[20] further pointed out that GLI1 substantially upregulated protein levels in colon cancer and colon tissue, and its expression was also helpful in predicting LNM, T stage, and postoperative metastasis-free survival, similar to our findings. PTTG1 induces significantly increased mRNA and protein expression in CRC tissues and cell lines, and its overexpression is closely related to LNM[21]. Moreover, the influence of PTTG1 on CRC cell migration and invasion is related to FOXM1 (forkhead box M1)[22], and its promoting effect on CRC cell growth and motility is associated with its induction by STAT3 (transcriptional activator 3)[14].

Analysis of the clinicopathological data showed that GLI1 and PTTG1 had no significant relationship with sex, age, tumor location, degree of tumor differentiation, and pathological stage but were significantly related to LNM. That is, patients with LNM had significantly higher expression rates of GLI1 and PTTG1. Similarly, we found significantly elevated GLI1 and PTTG1 expression in patients with LNM, suggesting that both GLI1 and PTTG1 are deeply involved in the pathological process of LNM in these patients. Yang et al[23] confirmed the significant correlation between GLI1 and distant metastasis in colorectal adenocarcinoma (CRAC) and clinical staging. Evidence has shown that GLI1-activated Hh signaling is strongly linked to poor clinicopathological status and low survival rate in patients with CRC, indicating the potential of GLI1 to predict poor clinicopathological characteristics and prognosis in patients with CRC[24]. GLI1 can also predict survival in patients with CRAC to some extent, with positive GLI1 expression indicating shorter survival. Besides indicating the presence of LNM, GLI1 can also predict the pathological status of patients with tumor size ≥5 cm and venous and nerve invasion[25]. However, the research on the correlation between PTTG1 and clinicopathological data in CRAC is limited. Another study also reported a significant positive correlation between PTTG1 protein levels and TNM staging in endometrial cancer but no significant correlation with other clinical parameters[26]. Furthermore, ROC analysis revealed that the AUC of both GLI1 and PTTG1 for predicting LNM was over 0.800, indicating that their expression can effectively predict LNM in patients undergoing radical surgery for CRC.

This study has several limitations. First, the single-center study design and small sample size would have affected the universality of the research results. Second, no basic investigation was conducted to explore the specific molecular mechanisms of GLI1, PTTG1, and LNM in CRC. Third, no analysis was performed to identify the risk factors that affect the efficacy of radical surgery or LNM in patients with CRC. Continuous optimization and supplementation should be implemented in future investigations to address the above deficiencies.

CONCLUSION

GLI1 and PTTG1 were aberrantly overexpressed in patients undergoing radical surgery for CRC. The expression of both genes was significantly correlated with LNM and can be used as predictors of LNM. Furthermore, GLI1 and PTTG1 expression has certain clinical implications on the evaluation of cases and the development of treatment plans.