Search

Background: Lung cancer is the leading cause of cancer-related death in most western countries in both, males and females, accounting for roughly 20-25% of all cancer deaths. For choosing the most appropriate therapy regimen a definite diagnosis is a prerequisite. However, histological characterization of bronchoscopic biopsies particularly with low tumor cell content is often challenging. Therefore, this study aims at (a) determining the value of DNA methylation analysis applied to specimens obtained by bronchoscopic biopsy for the diagnosis of lung cancer and (b) at comparing aberrantly CpG loci identified in bronchoscopic biopsy with those identified by analyzing surgical specimens., Results: We report the HumanMethylation450-based DNA methylation analysis of paired samples of bronchoscopic biopsy specimens either from the tumor side or from the contralateral tumor-free bronchus in 37 patients with definite lung cancer diagnosis and 18 patients with suspicious diagnosis. A differential DNA methylation analysis between both biopsy sites of patients with definite diagnosis identified 1303 loci. Even those samples were separated by the set of 1303 loci in which histopathological analysis could not unambiguously define the dignity. Further differential DNA methylation analyses distinguished between SCLC and NSCLC. We validated our results in an independent cohort of 40 primary lung cancers obtained by open surgical resection and their corresponding controls from the same patient as well as in publically available DNA methylation data from a TCGA cohort which could also be classified with high accuracy., Conclusions: Considering that the prognosis correlates with tumor stage at time of diagnosis, early detection of lung cancer is vital and DNA methylation analysis might add valuable information to reliably characterize lung cancer even in histologically ambiguous sample material.

Targeting checkpoint inhibitors using monoclonal antibodies results in significantly better outcome of cancer patients compared to conventional chemotherapy. However, the current companion diagnostics to predict response is so far suboptimal, since they base on more or less reliable immunohistochemical approaches. In order to overcome these limitations, we analyzed epigenetic modifications of PDCD1 (PD1), CD274 (PD-L1), and CTLA4 in NSCLC tissues from 39 patients. Results were correlated with transcriptome data. Significant differences in the CpG-methylation patterns between tumor tissues and matched controls were observed for CTLA4 and PDCD1 (PD1) showing a decreased methylation of these genes compared to matched tumor-free tissues from the same patients. Results were confirmed by bisulfide sequencing in an independent validation cohort. Hypomethylation also resulted in increased expression of these genes as shown by transcriptome data. These epigenetic pathways as a hallmark of NSCLC might be useful to generate more precise diagnostic approaches in the future.