Search

It has recently emerged that innate lymphocytes are more diverse than previously appreciated. In addition to natural killer cells, various subsets of innate lymphoid cells are now being characterized. It has become apparent that the transcriptional programs underlying lineage specification and cell fate decisions of innate lymphocytes strikingly resemble those of T cell subsets, suggesting that such transcriptional circuitry was already pre-formed in the evolutionary older innate immune system. Here, we will review recent advances in our understanding of the core transcriptional programs driving development and cell fate decisions of innate lymphocytes. We will also discuss whether these transcriptional programs are stable or flexible, thereby allowing for plastic adaptation of immune responses.

Introduction With the advent of molecular diagnostics and availability of targeted therapies, treatment of NSCLC has been improved, especially for tumours driven by mutant EGFR or ALK fusions. However, other frequently mutated genes, including KRAS, TP53 and STK11/LKB1, are still not targetable. Murine models of lung cancer that mimic human disease genomics and pathology can help refine diagnostic strategies, allowing discovery of new therapeutic options, identification of biomarkers and validation of the responses in vivo. Material and methods Primary cultures were derived from murine adenosquamous carcinoma (ASC) or adenocarcinoma (AC) tumours driven by mutant Kras and loss of Lkb1 (KL) or Tp53 (KP), and murine lung-derived healthy epithelia. Ex vivo drug sensitivity profiling, biomarker analysis and in vivo response validations were combined to identify NSCLC subtype-specific drug vulnerabilities. To identify biomarkers predictive of response, spatial analyses of murine and human tumours were conducted. Results and discussions By performing ex vivo functional screens, we identified histotype- and genotype-selective drug vulnerabilities. Inhibition of MEK, a downstream KRAS effector, caused a selective transient cytostasis in AC cells. However, long-term MEK inhibition was found to be ineffective in all cultures tested due to adaptive reactivation of MAPK and PI3K/AKT signalling. Resistance to MEK inhibitors was associated with activation of subtype-specific receptor tyrosine kinases: ERBB3 activation in KL;ASCs, adaptive ERBB2 and FGFR1 activation in KL;ACs, or adaptive FGFR1 activation in KP;ACs. Furthermore, while combined inhibition of MEK and ERBB kinases induced cell death in KL;ASC and KL;AC but not in KP;AC ex vivo culture, in vivo this combination showed cytotoxic response only in the squamous cell carcinoma (SCC) regions of KL;ASCs. This was in agreement with baseline ERBB activation occurring selectively in SCC tissue. Interestingly, activation of ERBB family receptors was detected in a significant proportion of human NSCLC tumours. Analysis of spatial signalling activities may therefore identify NSCLC patients that could benefit from combinatorial MEK and ERBB inhibition. Conclusion Our study validates the utility of NSCLC cultures to discover drug resistance mechanisms that can be overcome with subtype-specific drug combinations. Importantly, the combination of ex vivo drug response profiling and in situ spatial signalling shows that Kras-driven lung cancers stratify as functional subgroups.