Designing a novel 3D co-culture model incorporating patient T-cells to study lipid metabolism as a metabolic vulnerability in KRAS driven non-small cell lung cancer

Non-small cell lung cancer (NSCLC) accounts for 80-85% of all lung cancers worldwide, with KRAS-driven NSCLC representing 30% of these cases. Although immunotherapy has improved outcomes in lung cancer, only 27-46% of patients respond initially to immune checkpoint inhibitors (ICIs). One cohort of patients that do not respond to such treatment include those with KRAS-driven NSCLC harbouring a co-mutation in the LKB1 gene (KL), who have significantly poorer responses to immunotherapy compared to those with a KRAS and TP53 co-mutation (KP). However, the reason for this difference in response to immunotherapy is largely unknown. Recent studies demonstrate that metabolic vulnerabilities in KL genotypes can influence therapeutic response. We aim to identify and target these vulnerabilities in KL NSCLC, with the aim to enhance response to immunotherapy. Using the cBioPortal mRNA database and publicly available patient protein data, we completed a bioinformatic analysis of key genes, proteins, and metabolites involved in major metabolic pathways. Lipid metabolism displayed significant differences between KP and KL genotypes, specifically beta-oxidation and fatty acid synthesis. Following this, we conducted a drug screen using inhibitors targeting lipid metabolism. Fatty acid synthase (FASN) inhibitor, TVB-2640, showed the greatest efficacy in KL cells. Moreover, functional assays demonstrated reduced colony formation and migration following FASN inhibition in KL cells. Separately, we have successfully established novel 3D co-culture models of NSCLC cell lines and T-cells. This was achieved by embedding A549 NSCLC cells in a collagen gel matrix, with peripheral blood mononuclear cells (PBMCs) from patients with NSCLC enrolled on PLAN clinical trial (Trial ID: NCT05542485) added to the surrounding media. PBMCs were isolated, activated and expanded for 7 days using CD3, CD28 and IL-2. Following expansion, the T-cells were added to the A549 co-culture model for a further 4 days. These co-cultures were then paraffin embedded and stained using immunohistochemistry (IHC). Results demonstrated positive Ki67 staining and low cleaved caspase-3, indicating cell proliferation and minimal cell death. Moreover, IHC and flow cytometry analysis revealed 12% and 19% CD3 positive staining respectively, confirming successful T-cell infiltration in our co-culture model. In conclusion, we have identified fatty acid synthesis as a metabolic vulnerability in KL NSCLC. Targeting this pathway could be a potential therapeutic option for patients with this genotype of NSCLC. Moreover, we have developed a novel 3D co-culture model using NSCLC cells and patient T-cells. Further analysis, supported by our 3D co-culture model, aims to analyse the consequences of targeting fatty acid synthesis in combination with immunotherapy in the KL genotype of NSCLC.