Preview

PDF, English Download (7MB) | Terms of use

Abstract

Lung cancer remains the major cause of cancer related death worldwide. Recent developments in immunotherapy promise to be a gamechanger in early and advanced disease by overcoming the tumor immune escape. However, current immunotherapeutic approaches cannot be implemented for every patient, some lack a benefit and especially women often suffer from severe side effects or fail to respond to the treatment successfully. Glycodelin is a glycoprotein which is crucial for the establishment and maintenance of pregnancy. Glycodelin A, one of four glycosylation forms, suppresses the maternal immune system and allows the fertilized egg to invade into the decidual tissue. During pregnancy, it modulates the immune environment at the feto-maternal interface to prevent defense mechanisms towards the fetus as a semi-allograft. Interestingly, high glycodelin protein and corresponding progesterone-associated endometrial protein (PAEP) gene expression were also discovered in lung tumors compared to normal lung tissue. In the frame of this project I have investigated whether glycodelin in non-small cell lung cancer (NSCLC) shares the immunosuppressive function of glycodelin A known from pregnancy and could therefore serve as a novel target for future immunotherapies. For my study, I have used the primary cell lines 4950T and 170162T that secrete between 20-100 ng/ml glycodelin into the cell culture supernatant. The glycosylation structure was analyzed by lectin-based enrichment and could show that NSCLC derived glycodelin resembles the glycosylation pattern of glycodelin A isolated from amniotic fluid. The proteins shared high sialylation which is known to be crucial for immunosuppression. It is important to point out that sialic residues were detected weaker in glycodelin derived from 170162T, a NSCLC cell line that originates from a male patient. Thus, the protein might have sex-specific structures and functions. In vitro, NSCLC-derived glycodelin was bound and internalized by immune cells. It did not induce apoptosis but affected gene expression in monocytic and natural killer cells involved in tumor microenvironment and inflammation pathways. By using multiplex immunofluorescence and spatial analysis on 700 tissue samples, I could demonstrate that glycodelin binds to CD8+ T cells and CD163+ (M2) macrophages in tumor and stroma. Thus, glycodelin in NSCLC clearly interacts with surrounding immune cells and might modulate a pro-tumorigenic environment. The analysis of glycodelin in the serum of inoperable immunotherapy-treated NSCLC patients (n = 139) prior to immunotherapy showed that high serum concentrations of glycodelin were associated with a decreased progression-free survival (p < 0.001) of female patients receiving an anti-PD-1 / PD-L1 therapy. Glycodelin levels did not correlate with the hormones progesterone, estradiol, human chorionic gonadotropin (hCG), or testosterone in the serum. Consequently, it could serve as a predictive biomarker and enable better therapy decisions for female patients. As a first approach in vitro, I showed that glycodelin is targetable and that glycodelin binding to immune cells can be inhibited by using a monoclonal anti-glycodelin antibody. In conclusion, I could demonstrate that glycodelin has high potential of being a novel target in immuno-oncology and predictor of therapy response for NSCLC patients.