Associations Between Tumor Galectin-1 Expression And Glucose Metabolism And Their Prognostic Value In Patients With Lung Adenocarcinoma

Objective:Globally,lung cancer is the leading cause of cancer-related death among men and women.Lung adenocarcinoma is the most common histological subtype of NSCLC,the most common treatment strategies include surgery,chemotherapy and molecular targeted therapy.It remains a challenge that chemoresistance results recurrence and metastasis.Immunotherapy is giving new hope to patients with advanced lung adenocarcinoma.Immune checkpoint inhibitors have resulted in a remarkable paradigm shift in treatment options for NSCLC and are yielding promising clinical outcome.However,not all patients respond to immune checkpoint inhibitors because tumors may develop compensatory pathways to generate resistance.Consequently,there remains an unmet clinical need to identify new immune biomarkers and therapeutic targets in NSCLC.Galectin-1?Gal-1?has emerged as regulatory checkpoint that promotes tumor progression in many cancer types including lung cancer.Several Gal-1 targeting inhibitors have shown promising results in anti-cancer trials.Therefore,tumor Gal-1 status has potential as a predictive biomarker in patients with NSCLC.Immunohistochemistry is the gold standard in Gal-1 testing.However,the availability of sufficient quality and quantity of tissues for Gal-1 testing often remains a challenge.¹⁸F-FDG PET/CT,as a noninvasive imaging modality,has been demonstrated a correlation between FDG uptake and some biomarkers such as EGFR,PD-L1,Ki-67,VEGF and so on.However,the relationship between Gal-1 expression has not been assessed.The purpose of our study was to explore the association between tumor Gal-1 status and glucose metabolism and the prognostic value in lung adenocarcinoma.Methods:1.The medical records of 151 patients with primary lung adenocarcinoma who underwent surgery after ¹⁸F-FDG PET/CT at the First Affiliated Hospital of Dalian Medical University between June 2010 and August 2018 were retrospectively reviewed.The clinicopathological factors included age,gender,tumor location,tumor size,tumor differentiation,lymph node metastasis and tumor stage.Tumor Gal-1expression were examined by immunohistochemistry.The relationships between¹⁸F-FDG PET/CT metabolic parameters,including SUV_max,MTV and TLG,clinicopathological factors,and tumor Gal-1 expression were analyzed.Multivariate logistic regression analysis was used to identify predictors of tumor Gal-1 expression.ROC analysis was performed to estimate the optimal cutoff value of SUV_max.2.Tumor GLUT1,HK?,PFK,ALDOA,PKM2,LDHA,CD4,CD8,FOXP3 and CD68were examined by immunohistochemistry.Spearman correlation analysis and multivariate linear regression analysis were performed to identify the relationship between tumor Gal-1 expression and these key limiting enzymes and immune biomarkers expression.3.96 patients between June 2010 and October 2015 were followed up to understand the prognostic significance of ¹⁸F-FDG PET/CT metabolic parameters and tumor Gal-1 expression.Prognostic index includes PFS and OS.Survival analysis was performed using the Kaplan-Meier,Log rank test and Cox proportional hazard models.Results:1.Immunohistochemistry scores were used to stratify patients as Gal positive?n=70?or Gal-1 negative?n=81?.Tumor size was larger?Z=-3.929,P=0.000?and SUV_max,MTV,and TLG were significantly higher in patients with Gal-1 positive tumors compared to Gal-1 negative tumors?Z=-9.531,P=0.000;Z=-9.087,P=0.000;Z=-9.233,P=0.000?.Gal-1 positive tumors occurred more frequently in patients with lymph node metastasis,advanced stage tumors,and moderately or poorly differentiated carcinoma??²=45.542,P=0.000;?²=31.736,P=0.000;?²=25.546,P=0.000?.There was significant positive correlation between tumor Gal-1 and SUV_max,MTV and TLG?r=0.716,P=0.000;r=0.689,P=0.000;r=0.702,P=0.000?.SUV_maxax was the only independent predictor of tumor Gal-1 expression?P=0.000?.On ROC analysis,the optimal cutoff value of SUV_maxax for predicting tumor Gal-1expression was 5.1?AUC 0.937[95%CI 0.882-0.992];p<0.0001?,at which SUV_maxax predicted tumor Gal-1 expression with 95.2%sensitivity and 87.0%specificity.2.According to the cutoff value of immunohistochemistry scores,patients were divided into two groups,such as GLUT1 positive?n=75?or GLUT1 negative?n=76?,HK?positive?n=68?or HK?negative?n=83?,PFKL positive?n=65?or PFKL negative?n=86?,ALDOA positive?n=73?or ALDOA negative?n=78?,PKM2 positive?n=65?or PKM2 negative?n=86?,LDHA positive?n=62?or LDHA negative?n=89?.CD8⁺T cells and CD68⁺TAMs were detected both in cancer nest and stroma,CD4⁺T cells and FOXP3⁺Treg cells were mainly detected in cancer stroma.Spearman correlation analysis showed that there was significant positive correlation between tumor Gal-1and GLUT1,HK?,PFKL,ALDOA,PKM2,LDHA,FOXP3 and CD68?r=0.657,P=0.000;r=0.661,P=0.000;r=0.570,P=0.000;r=0.301,P=0.000;r=0.546,P=0.000;r=0.292,P=0.000;r=0.368,P=0.000;r=0.355,P=0.000?and there was significant negative correlation between tumor Gal-1 and CD8?r=-0.349,P=0.000?and there was no significant correlation between tumor Gal-1 and CD4?P>0.05?.Multivariate linear regression analysis demonstrated that only GLUT1,HK?,PFKL and CD8 correlated with Gal-1 expression?t=4.526,P=0.000;t=4.206,P=0.000;t=4.561,P=0.000;t=-4.251,P=0.000?.3.In all 96 patients,recurrence and metastasis were confirmed in33 patients,20 patients died at the last follow-up.Immunohistochemistry scores were used to stratify patients as Gal positive?n=42?or Gal-1 negative?n=54?;PFS and OS were significantly shorter in patients with Gal-1 positive tumors compared to those with Gal-1-negative tumors??²=38.690,P=0.000;?²=14.096,P=0.000?;PFS and OS were significantly shorter in patients with SUV_max?5.0,MTV?2.0 and TLG?7.2compared to those with SUV_max<5.0,MTV<2.0 and TLG<7.2??²=26.080,P=0.000,?²=8.261,P=0.004;?²=21.297,P=0.000,?²=6.958,P=0.008;?²=24.181,P=0.000,?²=6.448,P=0.011?.OS was significantly shorter in male??²=5.215,P=0.022?,patients with lymph node metastasis??²=24.051,P=0.000?and advanced stages??²=32.870,P=0.000?compared to those corresponding groups.PFS was significantly shorter in patients with moderately or poorly differentiated carcinoma??²=13.860,P=0.000?,lymph node metastasis??²=33.349,P=0.000?and advanced stages??²=38.915,P=0.000?compared to those corresponding groups.On multivariate analysis,tumor Gal-1 expression and advanced tumor stage?P=0.000,P=0.001?were independent prognostic indicators of poor PFS,while SUV_maxax and advanced tumor stage?P=0.024,P=0.000?were independent prognostic indicators of poor OS.Conclusion:Patients with Gal-1 positive tumor showed higher FDG uptake than those with Gal-1 negative,SUV_maxax was the only independent predictor of tumor Gal-1expression;There was significant positive correlation between tumor Gal-1 and GLUT1,HK?,PFKL,ALDOA,PKM2,LDHA,FOXP3 and CD68;There was significant negative correlation between tumor Gal-1 and CD8;Tumor Gal-1expression was independent prognostic indicator of poor PFS,while SUV_maxax was independent prognostic indicator of poor OS.