Serum galectin‐3 as a biomarker for screening, early diagnosis, prognosis and therapeutic effect evaluation of pancreatic cancer

Abstract Galectin‐3 plays an important role in cell‐cell adhesion, macrophage activation, angiogenesis, metastasis and apoptosis and is overexpressed in pancreatic cancer. We explored the importance of galectin‐3 in the screening, early diagnosis, prognosis and therapeutic effect evaluation of pancreatic cancer. A time‐resolved fluorescence immunoassay was performed to detect serum galectin‐3 level. Serum samples were collected from healthy controls and patients with pancreatic cancer before and after different treatments, and the relationships between galectin‐3 level and clinical parameters were analysed. Among the healthy controls, one individual with an abnormally high concentration of galectin‐3 (9.85 μg/L) was diagnosed with pancreatic cancer. Compared to the pre‐operative level, galectin‐3 concentration significantly decreased in patients with radical excision 1 month after surgery (P < .05), but showed no obvious change in patients who underwent palliative resection. Additionally, among patients with radical excision, carcinoma recurrence rate was significantly higher in those with increased or unchanged galectin‐3 level. Retrospective analysis revealed the extraordinarily high value and high specificity of galectin‐3 for predicting 3‐year survival (P < .001). Thus, galectin‐3 may serve as a potential biomarker for the screening and early diagnosis of pancreatic cancer and as an independent prognostic indicator in patients with pancreatic cancer.


| INTRODUC TI ON
Pancreatic cancer is a common malignant tumour of the digestive system. While the incidence of pancreatic cancer is increasing year by year, the associated mortality rate is predicted to rise and become the second highest among all malignant tumours by 2030. 1 Given the high degree of malignancy, poor prognosis, low survival rate and lack of clinical features, the diagnosis and treatment of pancreatic cancer are extremely difficult. At present, surgical resection combined with post-operative chemotherapy is the most effective regimen for pancreatic cancer treatment, but early diagnosis is still a serious concern. Tumour infiltration and distant metastasis may result in reduced surgical resection rate and poor response to chemotherapy. 2,3 Prognosis is greatly improved upon cancer diagnosis at an early operable stage. Until now, even recognized serum markers for pancreatic carcinoma, such as carcinoembryonic antigen (CEA) or carbohydrate antigen 19-9 (CA19-9), have not completely satisfied the criteria for early diagnosis. Therefore, it is imperative to develop more accurate and effective biomarkers for the early diagnosis, prognosis and therapeutic effect evaluation of pancreatic cancer.
Galectin-3 is an important member of the β-galactoside-binding protein family that is involved in cell-cell adhesion, cell-matrix interaction, macrophage activation, angiogenesis, metastasis and apoptosis. [4][5][6] Many studies have shown the expression of galectin-3 in various tumours, such as those of the gastrointestinal tract, cardiovascular, urinary and respiratory systems, breasts and thyroid gland. [7][8][9][10] Galectin-3 overexpression is closely related to the proliferation, invasion and malignancy of tumour cells. [11][12][13] In our previous proteomic study, 14,15 galectin-3 is overexpressed in pancreatic carcinoma tissues. A strong galectin-3 staining was predominantly observed in the cytoplasm and weak staining in the nucleus of tumour cells. Galectin-3 highly expressed in pancreatic cancer tissue was also found to be secreted into the peripheral blood. Serum galectin-3 level was higher in pancreatic cancer patients than in healthy individuals or patients with benign pancreatic lesions, periampullary tumours or acute pancreatitis. Galectin-3 may serve as a diagnostic marker for pancreatic cancer. However, the relationship between serum galectin-3 level and clinicopathological parameters remains elusive. Furthermore, we discussed the role of galectin-3 in the assessment of therapeutic effect and prognosis of pancreatic cancer. Cancer TNM classification of malignant tumours. 16   According to our previous study, 15 the diagnostic sensitivity for pancreatic cancer was 75.5% and the false-positive rate was 9.1% when the diagnostic threshold in TRFIA for serum galectin-3 detection was 3.77 μg/L.

| Statistical analysis
Statistical analysis was performed with SPSS 19.0. The chi-square test or Fisher's exact test was used for any four-dimensional table test. The measured levels of serum galectin-3 are presented as median (range). The rank-sum test was used to analyse the relationship between the groups and clinical parameters. Kaplan-Meier survival analysis was performed, and survival curves were obtained. Single clinical parameters and survival rates of patients with pancreatic cancer were analysed using the log-rank test. The effects of different parameters on patient survival were evaluated by a multivariate analysis using the Cox proportional hazard regression model. P < .05 was considered statistically significant.

| Serum galectin-3 levels in healthy controls
The concentrations of serum galectin-3 in 57 of 1850 healthy con-  Table S1.
Clinical test results of this patient with suspected pancreatic cancer were collected. No abnormalities were observed in serum tumour markers or routine blood and urine tests, except that total bilirubin level (21.3 μmol/L) was slightly elevated in the liver function examination. CT and MRI results revealed distal pancreatic duct expansion in the pancreatic neck space, possibly indicating pancreatic cancer ( Figure S1A-B). The patient was admitted to undergo radical resection in the general surgery department of our hospital.
Pathological analysis confirmed the diagnosis of moderately differentiated pancreatic carcinoma ( Figure S1C).  Table 1).

| Serum galectin-3 levels before and after radical resection and palliative resection
A total of 36 patients underwent surgical resection (radical resection, n = 21, palliative resection, n = 15). Galectin-3 levels before treatment and 1, 3 and 6 months after resection were detected to evaluate any relationship with the curative effect. No observable difference was noted between the two groups before treatment (Table S2, P > .05). Galectin-3 levels dramatically decreased 1 month after radical resection, and the difference was still statistically significant at 3 and 6 months post-operatively compared with pre-operative levels (Table S2, Figure 1A, P < .05). However, there was no remarkable difference between the three post-operative time-points (Table S2, P < .05). No significant change in galectin-3 levels was observed before and after palliative resection (Table S2, Figure 1B, P > .05).

| Survival analysis of patients undergoing surgical resection
Considering the 36 patients from the operative group, survival time was longer in patients who underwent radical resection than in those who underwent palliative resection (Kaplan-Meier analysis; Figure 2A, P < .05). Among the 21 patients who underwent radical resection, patients with reduced galectin-3 levels survived longer than those with increased or unchanged galectin-3 levels ( Figure 2B, P < .05).

| Comparison of serum galectin-3 levels before and after treatment in the non-operative group
Treatment was considered effective in 31 patients and ineffective in 35 of 66 patients evaluated 1 month after treatment (Table S3). No significant difference in galectin-3 levels was detected between the two groups of patients either before or after treatment. Further, galectin-3 levels did not significantly change after non-operative treatment (P > .05, Table S3, Figure S2).

| Correlation between median survival time and clinical parameters of patients with pancreatic cancer
For the 200 patients with pancreatic cancer who were followed up, the median survival time was 5.5 months. Survival time correlated with clinical parameters (pancreatic cancer patient age, serum galectin-3 level, CA19-9 and CEA levels, TNM stage, liver or lymph node metastasis, and treatment; P < .05, Table S4).

| D ISCUSS I ON
Poor prognosis and high mortality rate in patients with pancreatic cancer are largely associated with the lack of early diagnosis. The  level is thought to contribute to the poor prognosis of patients with pancreatic cancer. 22 The expression of interleukin-6 could also be used to predict the survival time of patients with pancreatic cancer with liver metastases. 23 Galectin-3 was overexpressed in pancreatic cancer tissues in our previous study. 14 In the present study, serum galectin-3 level was dynamically monitored in 36 patients, and no significant difference was observed between the two groups before treatment (P > .05). Compared to the pre-operative levels, serum galectin-3 levels in patients who underwent radical resection decreased 1 month after surgery (P < .05).  This study analysed the relationship between serum galectin-3 level and clinicopathological parameters, changes in galectin-3 concentration before and after treatment, and effects of galectin-3 on the prognosis of patients with pancreatic cancer. Our results suggest that galectin-3 is likely to become a biomarker for the screening and early diagnosis of pancreatic cancer and may serve as an independent prognostic indicator for patients with pancreatic cancer. However, the sample size of patients with pancreatic cancer in our study was not large, especially for each subgroup. Even fewer patients had undergone radical resection. Our results gave a great indication. Further studies with added number of cases are therefore warranted to confirm these results.

ACK N OWLED G EM ENT
This study was supported by grants from the Natural Science

CO N FLI C T O F I NTE R E S T
The authors have declared that no competing interest exists.

DATA AVA I L A B I L I T Y S TAT E M E N T
All data generated or analysed during this study are included in this article.