ANALYSE DU RÔLE DE LA VOIE NOTCH DANS LA RÉSISTANCE AU TRAITEMENT CIBLÉ DE KRASG12C DANS LE CANCER DU POUMON

DECIPHERING THE ROLE OF NOTCH PATHWAY IN RESISTANCE AGAINST KRASG12C TARGETED THERAPY IN LUNG CANCER

• Auteur / Autrice : Quentin Thomas
• Direction : Antonio Maraver, Xavier Quantin
• Type : Projet de thèse
• Discipline(s) : Biologie Santé
• Date : Inscription en doctorat le 09/01/2023
• Etablissement(s) : Université de Montpellier (2022-....)
• Ecole(s) doctorale(s) : Sciences Chimiques et Biologiques pour la Santé (Montpellier ; Ecole Doctorale ; 2015-....)
• Partenaire(s) de recherche : Laboratoire : IRCM - Institut de Recherche en Cancérologie de Montpellier
• Partenaire(s) de recherche : Equipe de recherche : Voies oncogéniques dans le cancer du poumon.

Mots clés

Mots clés libres

Cancer du poumon

Kras g12c

Notch

Lung Cancer

Kras g12c

Notch

Résumé

Description du projet scientifique Etat de l'art KRAS est parmi les oncogènes les plus fréquemment mutés dans les cancers humains et est associé à un mauvais pronostic et à une évolution agressive de la maladie, ce qui en fait l'un des principaux moteurs oncogéniques du cancer humain. Les mutations de KRAS sont le plus souvent trouvées dans les cancers épithéliaux tels que le cancer du pancréas, le cancer colorectal ou le cancer du poumon 1, ce dernier étant la principale cause de décès par cancer dans le monde avec 18,2 % de tous les cas de cancer 2. Dans le cancer du poumon non à petites cellules (NSCLC), KRAS est muté chez environ un tiers des patients, bien plus fréquemment que d'autres facteurs oncogènes tels que l'EGFR (15 %), les réarrangements ALK (5 %), les mutations MET (3 %), pour lesquels des thérapies ciblées sont disponibles 3. Pour KRAS, la grande majorité (83%) des mutations associées au cancer se produisent au codon 12, le variant G12C étant le plus fréquemment trouvé dans les CBNPC (11% de tous les patients) 1. En dépit des implications pertinentes des altérations de KRAS dans le cancer, les mutations de KRAS ont été reconnues comme non modifiables pendant de nombreuses années 4. Il est important de noter qu'après un effort exhaustif de ciblage des médicaments au cours des trois ou quatre dernières décennies, de nouveaux inhibiteurs covalents de KRASG12C ont été récemment développés dans le cadre d'essais cliniques et ont révélé des résultats prometteurs dans le CBNPC avec les inhibiteurs adagrasib 5 ou sotorasib 6. Grâce à ces résultats, le sotorasib a été récemment approuvé pour le traitement des patients atteints de NSCLC mutant KRASG12C en deuxième ligne par la FDA le 28 mai 2021, ce qui constitue la première thérapie ciblée approuvée pour les tumeurs présentant une mutation KRAS. Malheureusement, il existe un nombre important de patients mutés KRASG12C qui ne répondent pas du tout à ces molécules. La résistance aux médicaments anticancéreux peut être intrinsèque ou acquise. Étant donné l'absence de bénéfice chez environ 50-60% des patients 5 ;6, il est probable que certains sous-groupes de patients présentent une résistance intrinsèque aux inhibiteurs de KRASG12C. De plus, chez les patients qui répondent bien, une résistance acquise se développe inévitablement contre la thérapie ciblée, comme le montre la courte PFS obtenue 6;7. Il est donc important d'élucider complètement les mécanismes possibles de résistance intrinsèque et acquise aux inhibiteurs de KRASG12C. Ces connaissances aideront à stratifier les patients atteints de cancer piloté par KRAS pour qu'ils reçoivent des inhibiteurs de KRASG12C en monothérapie et à trouver un traitement combinatoire avec d'autres inhibiteurs de la voie unique. Le mécanisme précis de la résistance chez les patients atteints de cancer traités par un inhibiteur de KRASG12C n'est pas clair pour le moment. Il est probable que la résistance intrinsèque et la résistance acquise puissent coexister et s'entremêler chez le même patient traité par des thérapies ciblées. La biologie de la résistance acquise à l'inhibiteur de KRASG12C a été étudiée par des approches génomiques et des études fonctionnelles in vitro et in vivo dans l'espoir de trouver un moyen d'améliorer l'efficacité de ces nouvelles molécules. Les mécanismes de résistance putatifs identifiés comprennent des altérations sur la cible et au-delà de la cible et se répartissent en quatre classes différentes : i) les altérations de KRAS (mutations dans la poche de commutation II, autres mutations activatrices de KRAS, conversion de l'allèle cis de la mutation du codon KRASG12C, ou amplification de KRASG12C) ii) les mutations de contournement de KRASG12C dans la voie RAS/MAPK/PI3K (RTKs MET, EGFR, RET, ALK, FGFR3, BRAF, CRAF, MEK1, NRAS, PI3K), ii) transformation de l'état cellulaire d'un adénocarcinome à une histologie de carcinome épidermoïde et iv) activation de la signalisation hedgehog, Notch et WNT/b-caténine ainsi que des voies de transition épithélio-mésenchymateuse (EMT) et d'activation de l'angiogenèse tumorale 8-10. Parmi toutes ces possibilités, le projet actuel s'appuie sur la voie Notch. En effet, les travaux du laboratoire du Dr Maraver ont défini la voie Notch comme un acteur pertinent dans le NSCLC piloté par KRAS et ont montré un bénéfice thérapeutique lorsque les inhibiteurs de Notch étaient utilisés en monothérapie, un travail publié dans la revue Cancer Cell 11. Le laboratoire a récemment démontré que la voie Notch joue un rôle majeur dans la résistance aux inhibiteurs de la tyrosine kinase de l'EGFR dans les CBNPC dirigés par l'EGFR. Cette dernière étude, publiée dans le magazine Journal of Clinical Investigation 12, a ouvert la voie à des essais cliniques et a attiré l'attention de SpringWork Therapeutics, propriétaire du nirogacestat, un puissant inhibiteur de la g-sécrétase qui affecte la voie Notch et qui a montré des résultats très impressionnants dans une tumeur solide dépendante de Notch, à savoir la tumeur desmoïde 13. Un essai clinique de phase 1 commencera dès que possible à notre hôpital local, l'ICM (Montpellier). Scientific project and significance KRASG12C inhibitors have shown some efficacy in early clinical trials, however, responses have been variable and short, since intrinsic and acquired resistance limits their efficacy. The so far developed studies designed to elucidate the mechanisms of acquired resistance have proposed different combination therapies that could evade this kind of resistance and maximize their therapeutic potential. Nevertheless, these potential combination strategies should be tested in preclinical models derived from patients, which are the ones that best represent the intra- and inter-tumoral heterogeneity and best predict the clinical response to treatments and in GEMM that are immunocompetent and actually, combination of both type of preclinical models are the best choice to find new treatments. The purpose of this project is: firstly, expand the collection of preclinical models of KRAS-driven PDX, PDXDO. And secondly, to explore in our preclinical models, both PDX and PDXDO as well as GEMM, the efficacy of combining Notch inhibitors with KRASG12C inhibitors in order to avoid/delay the appearance of resistance against KRASG12C inhibitors as well as to regain KRASG12C inhibitors efficacy on those tumors that relapsed. Thus, we will contribute to personalize oncology selecting the right patient population that can benefit from each therapeutic strategy. Maintenant, nous voulons étendre notre propre collection de modèles PDX d'adénocarcinome pulmonaire piloté par KRAS et nous proposons de générer au moins 6 autres modèles. Pour générer les nouveaux modèles PDX, du matériel de résection ou de biopsie de tumeurs pulmonaires de patients sera obtenu en collaboration avec les départements d'oncologie et de pathologie, puis inoculé par voie sous-cutanée et développé chez la souris. Tous les patients fourniront un consentement éclairé écrit conformément à un protocole approuvé par le comité d'éthique local. Les expériences seront réalisées conformément aux directives européennes pour le soin des animaux de laboratoire, et le comité institutionnel de soin et d'utilisation des animaux approuvera cette étude au préalable. En bref, les PDX seront établis en implantant des fragments de tumeurs par voie sous-cutanée dans les deux flancs ou par voie orthotopique de souris nude (nu/nu). La taille de la tumeur sera mesurée une fois par semaine, et le volume de la tumeur sera calculé à l'aide de la formule ([longueur × largeur2]/2). Les tumeurs seront éliminées lorsque leur taille atteindra 1000mm3. Nous définirons les souris avec l'échantillon original du patient, comme le passage 0 (P0), tandis que les générations suivantes ont été numérotées consécutivement. Les modèles PDX seront considérés comme établis lorsque les tumeurs auront atteint le passage 2 (P2). Dans tous les passages, des fragments de tumeurs seront conservés pour des études histologiques et moléculaires ultérieures. Les neuf modèles PDX de poumon pilotés par KRAS dont nous disposons déjà sont caractérisés de manière exhaustive au niveau histologique, transcriptomique, génomique et protéomique. Les 6 nouveaux modèles seront caractérisés au même niveau histologique et moléculaire. Pour la caractérisation des nouveaux modèles : a) une caractérisation histologique sera effectuée en utilisant des microarrays de tissus (TMA), qui seront colorés automatiquement sur la plateforme BONDIII (Leica), en utilisant les réactifs recommandés par la société et les tranches seront quantifiées par des pathologistes. b) une caractérisation génomique et transcriptomique sera également effectuée. L'ADN et l'ARN seront extraits des PDX en P2. Ce matériel sera soumis au séquençage de l'exome entier (WES) et au séquençage du transcriptome entier (WTS). Enfin, nous utiliserons également le GEMM mentionné ci-dessus qui développe des adénocarcinomes pulmonaires ''de novo'' induits par KrasG12C 16. Nous suivrons la formation des tumeurs par petite imagerie, c'est-à-dire par tomographie assistée par ordinateur (CT) en utilisant l'appareil récent mais pleinement fonctionnel de l'IRCM (Montpellier). Après randomisation, nous traiterons les souris dès le début avec le véhicule (contrôle), le sotorasib ou l'adagrasib en traitement unique et les combinaisons de ces molécules avec le nirogacestat, inhibiteur de Notch, soit 5 groupes de traitement au total avec n = 10 souris par groupe. Nous suivrons la croissance tumorale à l'aide du CT et contrôlerons l'apparition de la résistance tumorale ainsi que la survie par analyse Kaplan- Meier. Le laboratoire du coordinateur a une grande expérience de l'utilisation de la tomodensitométrie et de la réalisation de courbes de survie chez la souris 11;12. Quant au GEEM, nous tirerons parti d'un sous-ensemble de souris qui développent une résistance au sotorasib dans le modèle de souris KrasG12C. A noter que le laboratoire du coordinateur a déjà montré que bien que les tumeurs répondent au sotorasib, après 3-4 mois toutes les tumeurs rechutent (données non publiées et non montrées par manque d'espace). Nous réalisons actuellement une expérience similaire avec l'adagrasib pour vérifier si c'est également le cas pour cet inhibiteur. Quoi qu'il en soit, nous allons co-traiter avec le nirogacestat les tumeurs qui rechutent à l'inhibiteur de KrasG12C en traitement unique afin de tester un regain de sensibilité au traitement, comme cela se produit dans les tumeurs pulmonaires dirigées par l'EGFR et résistantes aux TKI de l'EGFR lors de la combinaison des inhibiteurs de Notch 12. Bibliographie 1. Cerami, E., et al. The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov 2, 401-404 (2012). 2. Organization, W.H. https://www.who.int/news-room/fact-sheets/detail/cancer. (2021). 3. Pakkala, S. & Ramalingam, S.S. Personalized therapy for lung cancer: striking a moving target. JCI Insight 3(2018). 4. Ferrer, I., et al. KRAS-Mutant non-small cell lung cancer: From biology to therapy. Lung Cancer 124, 53-64 (2018). 5. Hallin, J., et al. The KRAS(G12C) Inhibitor MRTX849 Provides Insight toward Therapeutic Susceptibility of KRAS-Mutant Cancers in Mouse Models and Patients. Cancer Discov 10, 54-71 (2020). 6. Hong, D.S., et al. KRAS(G12C) Inhibition with Sotorasib in Advanced Solid Tumors. N Engl J Med 383, 1207-1217 (2020). 7. Janne, P.A., et al. Adagrasib in Non-Small-Cell Lung Cancer Harboring a KRAS(G12C) Mutation. N Engl J Med 387, 120-131 (2022). 8. Awad, M.M., et al. Acquired Resistance to KRAS(G12C) Inhibition in Cancer. N Engl J Med 384, 2382-2393 (2021). 9. Adachi, Y., et al. Epithelial-to-Mesenchymal Transition is a Cause of Both Intrinsic and Acquired Resistance to KRAS G12C Inhibitor in KRAS G12C-Mutant Non-Small Cell Lung Cancer. Clin Cancer Res 26, 5962-5973 (2020). 10. Tsai, Y.S., et al. Rapid idiosyncratic mechanisms of clinical resistance to KRAS G12C inhibition. J Clin Invest 132(2022). 11. Maraver, A., et al. Therapeutic Effect of gamma-Secretase Inhibition in Kras(G12V)- Driven Non-Small Cell Lung Carcinoma by Derepression of DUSP1 and Inhibition of ERK. Cancer Cell 22, 222-234 (2012). 12. Bousquet Mur, E., et al. Notch inhibition overcomes resistance to tyrosine kinase inhibitors in EGFR-driven lung adenocarcinoma. J Clin Invest 130, 612-624 (2020). 13. Kummar, S., et al. Clinical Activity of the gamma-Secretase Inhibitor PF-03084014 in Adults With Desmoid Tumors (Aggressive Fibromatosis). J Clin Oncol 35, 1561-1569 (2017). 14. Byrne, A.T., et al. Interrogating open issues in cancer precision medicine with patient- derived xenografts. Nat Rev Cancer 17, 254-268 (2017). 15. Guillen, K.P., et al. A human breast cancer-derived xenograft and organoid platform for drug discovery and precision oncology. Nat Cancer 3, 232-250 (2022). 16. Zafra, M.P., et al. An In Vivo Kras Allelic Series Reveals Distinct Phenotypes of Common Oncogenic Variants. Cancer Discov 10, 1654-1671 (2020). 17. Day, C.P., Merlino, G. & Van Dyke, T. Preclinical mouse cancer models: a maze of opportunities and challenges. Cell 163, 39-53 (2015

Description of the scientific project State of the art KRAS is among the most frequently mutated oncogene in human cancers and is associated with poor prognosis and an aggressive disease course, making it one of the top oncogenic drivers of human cancer. KRAS mutations are most frequently found in epithelial cancers such as pancreatic, colorectal or lung cancer 1, being the last one the leading cause of death by cancer worldwide with 18.2 % of all cancer cases 2. In non-small cell lung cancer (NSCLC), KRAS is mutated in around one third of patients, much more frequently than other oncogenic drivers such as EGFR (15%), ALK rearrangements (5%), MET mutations (3%), for which targeted therapies are available 3. For KRAS, the vast majority (83%) of cancer-associated mutations occur at codon 12, being G12C variant the most commonly found in NSCLC (11% of all patients) 1. Despite the relevant implications of KRAS alterations in cancer, KRAS mutations have been recognized as undruggable for many years 4. Importantly, after an exhaustive drug-targeting effort over the past 3–4 decades, novel KRASG12C covalent inhibitors, are being recently developed in clinical trials and have revealed promising results in NSCLC both with adagrasib 5 or sotorasib 6 inhibitors. With these results, sotorasib has been recently approved for the treatment of patients with second-line KRASG12C mutant NSCLC by the FDA on May 28, 2021, which marks the first approved targeted therapy for tumors with any KRAS mutation. Unfortunately, there is an important number of KRASG12C mutated patients that do not respond at all to these molecules. Resistance to anticancer drugs can be either intrinsic or acquired. Given the lack of benefit in about 50–60% of patients 5 ;6, it is likely that certain subgroups of patients present intrinsic resistance to KRASG12C inhibitors. Even more, inside the patients that respond well, acquired resistance inevitably develops against the targeted therapy, as show the short PFS obtained 6;7. Hence, it is important to fully elucidate possible intrinsic and acquired resistance mechanisms to KRASG12C inhibitors. This knowledge will help to stratify KRAS-driven cancer patients to receive KRASG12C inhibitors in monotherapy and to find combinatorial treatment with other singling pathway inhibitors. The precise mechanism of resistance in patients with cancer treated with KRASG12C inhibitor is unclear at the moment. It is likely that intrinsic and acquired resistance may co-exist and be intertwined in the same patient treated with targeted therapies. The biology behind KRASG12C inhibitor acquired resistance has been investigated with genome wide approaches and functional in vitro and in vivo studies in the hope of finding a way to improve the efficacy of these new molecules. Identified putative resistance mechanisms include on-target and beyond- target alterations and fall into four different classes: i) KRAS alterations (mutations in the switch II pocket, other activating KRAS mutations, cis allele conversion of the KRASG12C codon mutation, or KRASG12C amplification) ii) KRASG12C bypass mutations in the RAS/MAPK/PI3K pathway (RTKs MET, EGFR, RET, ALK, FGFR3, BRAF, CRAF, MEK1, NRAS, PI3K), ii) cell state transformation from adenocarcinoma to squamous cell carcinoma histology and iv) activation of hedgehog, Notch and WNT/b-catenin signaling as well as pathways for epithelial-mesenchymal transition (EMT) and tumor angiogenesis activation 8-10. Among all these possibilities, the current project relies on the Notch pathway. Indeed, work from the lab of Dr Maraver has defined the Notch pathway as a relevant actor in KRAS-driven NSCLC and showed therapeutic benefit when Notch inhibitors were used as monotherapy, a work published in Cancer Cell magazine 11. The lab recently demonstrated that the Notch pathway play a major role in resistance against EGFR tyrosine kinase inhibitors in EGFR-driven NSCLC. This last study, published in Journal of Clinical Investigation magazine 12, paved the way for clinical trials and attracted the attention of SpringWork Therapeutics, owners of nirogacestat, a potent g-secretase inhibitor that impinges the Notch pathway and showed very impressive results in a Notch dependent solid tumor, i.e., the desmoid tumor 13. A phase 1 clinical trial will start as soon as possible at our local hospital ICM (Montpellier). Other important part for the realization of this project would be the preclinical models used. It is important to highlight the urgently need of really predictive preclinical in vitro and in vivo models, that is, models that recapitulate the complexity of human tumors. Patient-derived xenografts (PDXs) have emerged as an important platform to elucidate new treatments and biomarkers in oncology. PDX models are used to address clinically relevant questions, including the contribution of tumor heterogeneity to therapeutic responsiveness, the patterns of cancer evolutionary dynamics during tumor progression and under drug pressure, and the mechanisms of resistance to treatment 14. Additionnally, it has been reported that matched PDXs and PDX-derived organoids (PDXDO) collections is feasible for drug screening and cost-effective for in vivo validation 15. We will also use a genetic engineered mouse model (GEMM) that upon cre recombinase induction in the lung develops lung adenocarcinoma dependent on KRASG12C mutation 16. Of note, this mouse model is very similar to the inducible KRASG12V mouse model widely used by Dr Maraver as the study in Cancer Cell magazine witness 11. Scientific project and significance KRASG12C inhibitors have shown some efficacy in early clinical trials, however, responses have been variable and short, since intrinsic and acquired resistance limits their efficacy. The so far developed studies designed to elucidate the mechanisms of acquired resistance have proposed different combination therapies that could evade this kind of resistance and maximize their therapeutic potential. Nevertheless, these potential combination strategies should be tested in preclinical models derived from patients, which are the ones that best represent the intra- and inter-tumoral heterogeneity and best predict the clinical response to treatments and in GEMM that are immunocompetent and actually, combination of both type of preclinical models are the best choice to find new treatments. The purpose of this project is: firstly, expand the collection of preclinical models of KRAS-driven PDX, PDXDO. And secondly, to explore in our preclinical models, both PDX and PDXDO as well as GEMM, the efficacy of combining Notch inhibitors with KRASG12C inhibitors in order to avoid/delay the appearance of resistance against KRASG12C inhibitors as well as to regain KRASG12C inhibitors efficacy on those tumors that relapsed. Thus, we will contribute to personalize oncology selecting the right patient population that can benefit from each therapeutic strategy. Quality and originality of the project The originality of our project is based on discoveries and tools, including: state-of the art clinical patient-derived xenografts resistant to KRASG12C therapy, technology to follow in a single cell manner the response against targeted therapy, as well as recent acquisition of a genetic engineered mouse model with inducible KRASG12C to work also in an immunocompetent background. Objectives 1. Exploring combination therapies based on KRASG12Ci and NOTCHi in primary sensible and intrinsic resistant preclinical models. 2. Exploring combination therapies based on KRASG12Ci and NOTCHi in acquired resistant preclinical models. Description of the activities, expenses in the framework of the project and timetable 1. Exploring combination therapies based on KRASG12Ci and NOTCHi in primary sensible and intrinsic resistant preclinical models. 1.a Extension and characterization of our own collection of PDX, PDXDO and GEMMs of KRAS- driven cancer. A comprehensively characterized collection of 45 NSCLC PDX models which contain 15 models of adenocarcinomas, 8 of them with KRAS mutations and one with KRAS gain (under review in Clinical Cancer Research) and a collection of PDXDO matched has already been generated. All PDXs retain the same morphology, expression of different biomarkers and presence of different genomic alterations from their corresponding original patient tumor and across the passages (unpublished data not shown due to lack of space). We have verified that 100% of the PDX generates PDXDO, which maintain the same morphology, expression of markers and driver mutations and in the same allele frequency than the PDX, and therefore than the patient tumor. In addition, we have verified that both PDX and PDXDO collections are good preclinical tools to evaluate therapies, since they predict very well the response found in clinic. Now, we want to expand our own collection of lung adenocarcinoma KRAS-driven PDX models and we propose to generated at least another 6. For generating the new PDX models, resected or biopsy material from lung tumors from patients will be obtained by collaboration with Oncology and Pathology departments and inoculated subcutaneously and expanded in mice. All patients will provide written informed consent according to a protocol approved by the local ethics committee. The experiments will be performed according to the European guidelines for laboratory animal care, and the Institutional Animal Care and Use Committee will approve this study previously. Briefly, PDXs will be established by implanting fragments of tumors subcutaneously in both flanks or orthotopically of nude (nu/nu) mice. Tumor size will be measured once a week, and tumor volume calculated using the formula ([length × width2]/2). Tumors will be passaged when its size reached 1000mm3. We will define the mice with the original patient sample, as passage 0 (P0), whereas subsequent generations were numbered consecutively. PDX models will be considered established when the tumors achieved passage two (P2). In all passages, tumor fragments will be conserved for subsequent histological and molecular studies. The nine KRAS-driven lung PDX models we already have, are comprehensively characterized at the histological, transcriptomic, genomic and proteomic level. The 6 new ones will be characterized at the same histological and molecular level. For characterizing the new models: a) a histological characterization will be performed using tissue microarrays (TMA), which will be stained automatically on the BONDIII platform (Leica), using the reagents recommended by the company and the slices will be quantified by pathologists. b) a genomic and transcriptomic characterization will be also performed. DNA and RNA will be extracted from PDXs in P2. This material will be subjected to Whole Exome Sequencing (WES) and Whole Transcriptome Sequencing (WTS). In this project, we will develop PDXDO from every PDX (n = 15, 9 old and 6 news) in order to have an in vitro patient-derived tool in which explore combination therapies based on KRASG12Ci and NOTCHi. For that, fresh tumor tissue from PDX will be processed for organoid culture. Briefly, the tumor will be minced into small fragments and treated for 1-2h with digestion solution (DMEM/F12 supplemented with HEPES, GlutaMax, FBS, Primocin, collagenase IA and hialuronidase). Red blood cells will be removed with cell lysis buffer. The pellet will be resuspended in Matrigel (20000-40000 cells in 50μl of Matrigel Growth Factor Reduced). 50μl matrigel domes will be plated into a pre-warmed 24-well plate. Once Matrigel is solidified, Media will be added (DMEMF12, HEPES1%, glutamax1%, primocin1%, B-27 supplement 1X, 10.5uM RhoKi, 2%FBS, 3ng/ml EGF, 5ug/ml Insulin, 1ug/ml Hydrocortison). The plate will be placed in 37oC incubators and culture media will be changed every four days. The PDXDO will be characterized histopathologically and molecularly in a similar manner as the PDX (see above). 1.b Exploring combination therapies based on KRASG12Ci and NOTCHi in primary sensible and intrinsic resistant preclinical models in vitro and in vivo. We will combine both sotorasib or adagrasib with nirogacestat (Notch inhibitor). These combinations will first be tested in vitro using the collections of PDXDO with the intention of finding the most effective combination for each model and thus define, according to their genomic and transcriptomic background, which therapy would be best for each patient. The most effective therapies will be validated in vivo by treating their corresponding PDX model and GEMMs. For the in vitro efficacy experiments: cells will be resuspended in Matrigel Growth Factor Reduced (Corning) and 5000 cells will be seeded in 96-well plates in 36ul of Matrigel per well. After solidifying the Matrigel, 100ul of culture medium (objective 1a) will be added to each well. Cells will be incubated O/N at 37oC and drugs will be added and treated for 1 week. We will test three concentrations of each drug and their combinations. Proliferation will be measured using CellTiter-Glo Cell Viability Assay (Promega) according to manufacturer's instructions. For the in vivo efficacy experiments: PDX tumors will be measured twice per week after implantation. When tumor volumes reached 200 mm3, mice will be divided into 4 groups (vehicle, drugs in monotherapies and drugs in combination) of 6-8 mice each. The treatments will last 3-6 weeks and we will collect tumor sample after short and long treatments in order to analyzed the activation of different related signaling pathways by transcriptomic and proteomic analysis. We will use this analysis in order to understand the action mechanisms of the combinations. Finally, we will use also the above-mentioned GEMM that develops “de novo” KrasG12C-driven lung adenocarcinomas 16. We will follow tumor formation by small imaging, i.e., computed tomography (CT) using the recent but fully functional device at IRCM (Montpellier). After randomization, we will treat mice from the beginning with vehicle (control), sotorasib or adagrasib in single treatments and the combinations of those molecules with the Notch inhibitor nirogacestat, i.e., 5 treatment groups in total with n = 10 mice per group. We will follow tumor growth using the CT and monitor tumor resistance appearance as well as survival by Kaplan- Meier analysis. Lab of coordinator has plenty of experience both using CT and performing mouse survival curves 11;12. 2. Exploring combination therapies based on KRASG12Ci and NOTCHi in acquired resistant preclinical models. Treatments will be started in mice bearing tumors of 400mm3 size. When tumor volume 3 decreased below 200mm, treatment is interrupted, and reinitiated when tumors re-grew, i.e., after reaching 400 mm3 volume. Mice will be killed when the tumor size surpassed a diameter of 1000 mm3 or mouse weight decreased by 20% (i.e., upon reaching the human end-point). Tumors will be then excised, cut in pieces and re-implanted into new host mice (P2) in which treatment will be performed following the same criteria. When a given treatment will show no effect on any given tumor, we will amplify the resistant model in order to freeze enough material to characterize the model and to evaluate new therapeutic strategies in more mice (as explain above). These models will be treated with the Notch inhibitor nirogacestat in combination with either sotorasib or adagrasib to test for regain in sensitivity to these treatments, as the lab did before with EGFR TKIs in EGFR-driven lung cancer 12. As for the GEEM, we will take advantage of a subset of mice that develop resistance against sotorasib in the KrasG12C mouse model. Of note, lab of coordinator already showed that although tumors respond against sotorasib, after 3-4 months all tumors relapse (unpublished data and not shown due to lack of space). We are at the moment performing a similar experiment with adagrasib to test if this is also the case for this inhibitor. In any case, we will co- treat with nirogacestat those tumors relapsing to the KrasG12C inhibitor as single treatment to test for a regain in sensitivity to the treatment as it happens in EGFR-driven lung tumors resistant to EGFR TKIs upon combination of Notch inhibitors 12. Bibliography 1. Cerami, E., et al. The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov 2, 401-404 (2012). 2. Organization, W.H. https://www.who.int/news-room/fact-sheets/detail/cancer. (2021). 3. Pakkala, S. & Ramalingam, S.S. Personalized therapy for lung cancer: striking a moving target. JCI Insight 3(2018). 4. Ferrer, I., et al. KRAS-Mutant non-small cell lung cancer: From biology to therapy. Lung Cancer 124, 53-64 (2018). 5. Hallin, J., et al. The KRAS(G12C) Inhibitor MRTX849 Provides Insight toward Therapeutic Susceptibility of KRAS-Mutant Cancers in Mouse Models and Patients. Cancer Discov 10, 54-71 (2020). 6. Hong, D.S., et al. KRAS(G12C) Inhibition with Sotorasib in Advanced Solid Tumors. N Engl J Med 383, 1207-1217 (2020). 7. Janne, P.A., et al. Adagrasib in Non-Small-Cell Lung Cancer Harboring a KRAS(G12C) Mutation. N Engl J Med 387, 120-131 (2022). 8. Awad, M.M., et al. Acquired Resistance to KRAS(G12C) Inhibition in Cancer. N Engl J Med 384, 2382-2393 (2021). 9. Adachi, Y., et al. Epithelial-to-Mesenchymal Transition is a Cause of Both Intrinsic and Acquired Resistance to KRAS G12C Inhibitor in KRAS G12C-Mutant Non-Small Cell Lung Cancer. Clin Cancer Res 26, 5962-5973 (2020). 10. Tsai, Y.S., et al. Rapid idiosyncratic mechanisms of clinical resistance to KRAS G12C inhibition. J Clin Invest 132(2022). 11. Maraver, A., et al. Therapeutic Effect of gamma-Secretase Inhibition in Kras(G12V)- Driven Non-Small Cell Lung Carcinoma by Derepression of DUSP1 and Inhibition of ERK. Cancer Cell 22, 222-234 (2012). 12. Bousquet Mur, E., et al. Notch inhibition overcomes resistance to tyrosine kinase inhibitors in EGFR-driven lung adenocarcinoma. J Clin Invest 130, 612-624 (2020). 13. Kummar, S., et al. Clinical Activity of the gamma-Secretase Inhibitor PF-03084014 in Adults With Desmoid Tumors (Aggressive Fibromatosis). J Clin Oncol 35, 1561-1569 (2017). 14. Byrne, A.T., et al. Interrogating open issues in cancer precision medicine with patient- derived xenografts. Nat Rev Cancer 17, 254-268 (2017). 15. Guillen, K.P., et al. A human breast cancer-derived xenograft and organoid platform for drug discovery and precision oncology. Nat Cancer 3, 232-250 (2022). 16. Zafra, M.P., et al. An In Vivo Kras Allelic Series Reveals Distinct Phenotypes of Common Oncogenic Variants. Cancer Discov 10, 1654-1671 (2020). 17. Day, C.P., Merlino, G. & Van Dyke, T. Preclinical mouse cancer models: a maze of opportunities and challenges. Cell 163, 39-53 (2015).