Transcriptome and protein network analyses of 3D-tissue lung cancer models reveal combinatorial targets for...

Detailed analysis of RNAseq datasets combined with signaling networksin physiologically relevant models are fundamental steps in the fight against drug resistance. For this purpose, we compare RNAseq data of different responsive KRAS^G12C-mutant cell lines, H358 and HCC44, to illustrate distinct resistance scenarios against a KRAS-inhibitor. We apply the following steps to investigate this: (I) Analysis of variance shows several differentially expressed marker genes for non-small cell lung cancer (NSCLC) and confirms HCC44 cells to be more aggressive. (II) Protein network analysis of key players of resistance reveals upregulation of Dachshund homolog 1 (DACH1) only in treated H358 cells. (III) A systematic analysis of NSCLC with a semi-quantitative signaling network predicts several protein targets counter-acting the KRAS^G12C-mutation, including DACH1. (IV) We show correlations between the expression level of these genes and resulting survival outcomes using TCGA data from patients. (V) We validate key signature genes by quantitative PCR in H358 cells. Our study identifies DACH1 as marker for resistance from RNAseq data of a 3D tissue NSCLC model validated with patient survival data. We predictmost promising candidates for combination therapies in silico.