High-throughput analyses for the identification of progression-specific factors in non-small-cell lung cancer

Lung cancer (LC) is the leading cause of cancer-related death worldwide, and clinical outcome is highly dependent on the stage at diagnosis. High mortality rates are mainly caused by patients diagnosed with already distant diseases. Histologically, LC is divided into non-small-cell lung cancer (NSCLC) and small-cell lung cancer (SCLC), accounting for 85 % and 15 % of cases, respectively. Although significant advances in treating advanced lung cancer have been achieved over the past decades, the overall cure and survival rates remain low. Hence, a deeper understanding of molecular factors contributing to lung cancer progression and relapse might pave the way for new treatment strategies and improved patient outcomes. For this purpose, we used high-throughput methods for the identification and mechanistic analysis of metastasis-modulating factors derived from an unbiased in vivo shRNA screen. From the top targets, we selected the aldehyde dehydrogenase Aldh9a1, a crucial enzyme involved in the biosynthesis of the fatty acid transporter carnitine, for further characterization of its potential metastasis-suppressive function. The suppression of Aldh9a1 in two adenocarcinoma cell lines (NCI-H1975 and A549) induced a mesenchymal shift along with increased invasive capacities and resistance to anoikis in vitro. In addition, major metabolic pathways were transcriptionally altered upon ALDH9A1 knock-down, highlighting the impact of Aldh9a1 on cell metabolism. Our results suggest a link between fatty acid oxidation (FAO) and metastatic properties in our lung models, which is not observed in other cancer types. Further insights into the underlying metabolic programs might offer new insights into tumor-type specific FAO regulation that could potentially be exploited for novel treatment options. Besides metastasis giving rise to secondary lesions, tumor relapse due to intratumoral heterogeneity and clonal selection contributes to poor clinical outcomes. Therefore, we developed a computational workflow to identify extrachromosomal circular DNA (ecDNA) from long-read sequencing data. As ecDNAs are massive drivers of intratumoral heterogeneity correlating with poor patient outcomes in various cancers, their presence in tumor cells serves as a negative prognostic marker. Upon read mapping against the human genome, circular graphs were generated and tested for plausibility. We demonstrated the robustness of our pipeline by recovering mitochondrial genomes from circular-enriched lung cancer samples and recovered the 966 kb extrachromosomal MYCN oncogene amplicon derived from the neuroblastoma cell line Kelly, which served as a positive control. Additionally, our pipeline provides an experimental validation step of predicted circles by amplifying the circle junction using polymerase chain reaction (PCR). Our pipeline allows for the identification and structural analysis of ecDNAs and additionally might facilitate longitudinal tracking of ecDNAs during a patient’s treatment course via PCR as an early-warning system for treatment failure and disease recurrence.