Non-small cell lung cancer (NSCLC) is a heterogeneous disease with unique combinations of somatic molecular alterations in individual patients, as well as significant differences in populations across the world with regard to mutation spectra and mutation frequencies. Here we aim to describe mutational patterns and linked clinical parameters in a population-based NSCLC cohort. Materials and methods: Using targeted resequencing the mutational status of 82 genes was evaluated in a consecutive Swedish surgical NSCLC cohort, consisting of 352 patient samples from either fresh frozen or formalin fixed paraffin embedded (FFPE) tissues. The panel covers all exons of the 82 genes and utilizes reduced target fragment length and two-strand capture making it compatible with degraded FFPE samples. Results: We obtained a uniform sequencing coverage and mutation load across the fresh frozen and FFPE samples by adaption of sequencing depth and bioinformatic pipeline, thereby avoiding a technical bias between these two sample types. At large, the mutation frequencies resembled the frequencies seen in other western populations, except for a high frequency of KRAS hotspot mutations (43%) in adenocarcinoma patients. Worse overall survival was observed for adenocarcinoma patients with a mutation in either TP53, STK11 or SMARCA4. In the adenocarcinoma KRAS-mutated group poor survival appeared to be linked to concomitant TP53 or STK11 mutations, and not to KRAS mutation as a single aberration. Similar results were seen in the analysis of publicly available data from the cBioPortal. In squamous cell carcinoma a worse prognosis could be observed for patients with MLL2 mutations, while CSMD3 mutations were linked to a better prognosis. Conclusion: Here we have evaluated the mutational status of a NSCLC cohort. We could not confirm any survival impact of isolated driver mutations. Instead, concurrent mutations in TP53 and STK11 were shown to confer poor survival in the KRAS-positive adenocarcinoma subgroup.

BackgroundIn recent years, nucleotide sequencing has become increasingly instrumental in both research and clinical settings. This has led to an explosive growth in sequencing data produced worldwide. As the amount of data increases, so does the need for automated solutions for data processing and analysis. The concept of workflows has gained favour in the bioinformatics community, but there is little in the scientific literature describing end-to-end automation systems. Arteria is an automation system that aims at providing a solution to the data-related operational challenges that face sequencing core facilities.FindingsArteria is built on existing open source technologies, with a modular design allowing for a community-driven effort to create plug-and-play micro-services. In this article we describe the system, elaborate on the underlying conceptual framework, and present an example implementation. Arteria can be reduced to 3 conceptual levels: orchestration (using an event-based model of automation), process (the steps involved in processing sequencing data, modelled as workflows), and execution (using a series of RESTful micro-services). This creates a system that is both flexible and scalable. Arteria-based systems have been successfully deployed at 3 sequencing core facilities. The Arteria Project code, written largely in Python, is available as open source software, and more information can be found at https://arteria-project.github.io/ .ConclusionsWe describe the Arteria system and the underlying conceptual framework, demonstrating how this model can be used to automate data handling and analysis in the context of a sequencing core facility.

No abstract