Lipocalin 2 – Devil or Angel? A pan cancer study.

Abstract

e15609 Background: Lipocalin2 (LCN2, also known as neutrophil gelatinase-associated lipocalin), is a protein that in humans is encoded by the LCN2 gene. It’s abnormal expression serves critical roles in the EMT transition process, angiogenesis, cell migration and invasion in many cancers. We aim to study the expression of LCN2 in tumours with respect to normal and its association with survival in multiple forms of cancer. Methods: On analysis of 23 TCGA datasets containing gene expression data for both tumour and adjacent normal samples, normalized RNAseq RSEM values of LCN2 were compared and differences between median expression levels were assessed using Wilcoxon rank sum test. Effect of LCN2 expression (lowest and highest tertiles) in tumour samples were estimated using Kaplan-Meier survival model. Correlation of LCN2 expression with immune cell population estimated using MCP Counter tool in Colonic Adenocarcinoma (CAC) and Head and Neck Squamous Cell Carcinoma (HNSCC) dataset were assessed. Results: LCN2 overexpression in tumour over normal was seen in 11 cancers viz. CAC, Cholangiocarcinoma, Esophageal ca, Kidney chromophobe, Kidney renal papillary cell ca, HCC, Lung adenoca, Rectum adenoca, Skin cutaneous melanoma, Thyroid ca and endometrial ca whereas LCN2 under-expression in tumour over normal was observed in 4 including HNSCC, Breast invasive ca, RCC and Prostate adenoca. For 8 cancers (Urothelial ca, Cervical and endocervical ca, Glioblastoma, Lower grade glioma, Lung squamous cell ca, Pancreatic adenoca, Phenochromocytoma and Stomach adenoca) no significant difference was observed between normal and tumour samples. LCN2 expression shows hazardous effect on OS in Renal Clear Cell ca (HR = 2.38, p = 0.00027), Glioblastoma (HR = 1.57, p = 0.036) and Pheochromocytoma (HR = 8.03, p = 0.03). On the other hand, high expression of LCN2 shows very mild protective effect on tumour progression in HNSCC (HR = 0.71, p = 0.05) and Prostate ca (HR = 0.46, p = 0.06). In CAC, the normal samples show a correlation between LCN2 expression and T-cells (Pearson r = 0.45, p = 0.0028) and with the T-cell chemo attractant CXCL10 (Pearson r = 0.5, p < 0.0001). Such correlations are broken in tumour samples. In HNSCC no correlations are visible between LCN2 expression and immune cell population estimate in either tumour or normal samples. Conclusions: Literature shows LCN2 as a potential therapeutic target. However, initial analysis of TCGA data fails to establish its role as pro- or anti-tumorigenic in multiple cancers. Further tissue-specific analysis including in vitro and invivo studies are needed to confirm its role in cancer.