Targeting cell state transitions driving resistance to KRAS inhibitors in lung cancer

Lung adenocarcinoma (LUAD), commonly driven by KRAS mutations, is responsible for 7% of all cancer mortality. LUAD predominantly arises from alveolar type 2 (AT2) cells, which function as facultative alveolar stem cells by self-renewing and replacing alveolar type 1 (AT1) cells. The first allele-specific KRAS inhibitors (KRASi) were recently approved in LUAD, but clinical benefit is limited by intrinsic and acquired resistance. Histologic transformation of LUAD to squamous cell cancer (adenosquamous transition, AST) has been reported as a form of acquired resistance to KRASi. Such cancer cell state transitions are emerging as central mechanisms of resistance to oncologic therapies. Importantly, they are not associated with acquired genetic events but are driven by epigenetic reprogramming and transcriptional re-wiring of cellular differentiation programs. To meet the challenge of treatment resistance in the clinic, identifying and targeting cell states that confer resistance to KRAS inhibitors must be sought with urgency. We used genetically engineered mouse models, patient-derived xenografts and organoids, and patient samples to identify LUAD differentiation states that drive resistance to KRASi. We found KRASi promote a quiescent AT1-like cancer cell state in LUAD tumors. The AT1-like LUAD cells exhibit high growth potential upon treatment cessation, whereas ablation of the AT1-like cells sensitizes tumors to KRASi and robustly improves treatment response. Notably, targeting KRAS in LUAD tumors harboring mutations in STK11, encoding the LKB1 tumor suppressor, invariably leads to histologic transformation into AST that is resistant to KRASi. These findings implicate genetically deterministic differentiation programs as key drivers of resistance to KRASi in lung cancer, whereby LKB1-proficient cancer cells adopt an AT1-like state and LKB1-deficient cells undergo a squamous transformation. We hypothesize targeting either the AT1-like or the squamous transition state, informed by LKB1status, will overcome resistance to KRASi in LUAD. To address this hypothesis, we will (i) investigate the origin and fate of the alveolar/squamous states in tumors using lineage- tracing; (ii) model outcomes of cytoablative therapies targeting the AT1-like/squamous states utilizing lineage- ablation; (iii) uncover molecular cell state drivers and evaluate them as targets in the context of LKB1 proficiency or deficiency. To do this, we will take advantage of novel genetically engineered mouse models (GEMMs) of KRAS(G12D) or KRAS(G12C) mutant LUAD that we have developed. We validate our findings in human patient- derived tissues and xenografts. Our proposed study will allow us to establish the AT1-like and squamous transition states (i) as LKB1 genotype-specific biomarkers of intrinsic resistance and relapse potential in LUAD and (ii) as targets for cytoablative therapies to overcome resistance to KRASi, as well as to identify molecular drivers of these cell states. Thus, this work will elucidate novel cellular and molecular mechanisms of response to KRAS-targeted therapy and inform strategies to overcome treatment resistance in LUAD patients.