MetLung, an innovative 3D lung metastasis model

Metastasis is the leading cause of death for cancer patients. The most common location in pediatric solid tumor patients is the lung. Our failure to cure patients with metastasis is due to a lack of knowledge on the metastatic niche, resulting from scarcity of relevant tumor models. Animal models are still a gold standard in preclinical research, although they often fail to recapitulate human biology and poorly predict patient response to drugs. Recent technological advances propel alternative approaches: Organoids are established in vitro from tissue stem cells and mimic architectural and functional characteristics of their corresponding in vivo organ, while 3D-bioprinted constructs can mimic the physical and chemical properties of the tissue microenvironment. We aim at constructing versatile and scalable in vitro alternatives to animal models of solid tumor lung metastasis for pre-clinical drug testing. The models will be designed in a stepwise fashion from less to more complex. To achieve the best possible in vitro approximation of the metastatic site we will inform the design of our models by analyzing actual lung metastases of various pediatric solid tumors by single-cell and spatial genomics. Cellular complexity and physical properties of the lung niche will be simulated through airway organoids and 3D-printed constructs. The models will be populated with pediatric patient- derived cells from the metastatic site and used to screen for drug sensitivities. This way, we will provide proof of concept for patient-specific 3D-models of lung metastatic pediatric tumors as an alternative to animal studies to guide personalized drug selection for patients with advanced disease. For the first time, state-of-the-art single cell genomics, organoid and 3D- bioprinting technologies will be combined to mimic tumor growth in its metastatic niche in vitro. Through specific targeting of the tumor/niche interactions we aim to develop more efficient, biology-based, personalized treatments for lung metastatic disease.

The most frequent cause of death of cancer patients is the development of distant metastases. Many solid tumors preferentially metastasize to the lung, among them bone sarcomas, which most frequently affect children and adolescents. To develop therapies that efficiently target lung metastatic disease, adequate models are needed that allow tracking of tumor cells in the lung context. So far, anti-metastatic drug testing was only possible in animal models, which only partially recapitulate human biology. In this project, we developed a modular purely human organoid-based culture system that allows to recapitulate tumor cell infiltration into the lung ex vivo. This way we identified distinct lung infiltration patterns of Ewing sarcoma and osteosarcoma, the most frequent bone sarcomas in young patients. While osteosarcoma cells penetrated deep into the lung organoids, Ewing sarcoma cells remained closer to the organoid surface. The systematic review of computer tomography images of lung metastases in a series of bone sarcoma patients resulted in the identification of similar tumor infiltration patterns, validating our ex vivo results: osteosarcoma metastases were generally found deep in the lung parenchyma while Ewing sarcoma metastases remained closer to the lung periphery. The adaptation of our model to liquid culture conditions enabled us to explore early reactions of the lung epithelial cell types to tumor cell contact. Phenotypically, we observed an early and pronounced fibrotic structural change in lung organoid appearance. Single-cell transcriptome sequencing revealed molecular alterations in several lung epithelial cell types consistent with a tissue repair response to tumor cell contact. To validate these results in patients, we explored tissue sections of 21 lung metastases in comparison to their respective primary bone tumors by spatial transcriptomics technology. These studies confirmed the appearance of a conserved tissue repair gene expression signature in the immediate vicinity of the metastatic lesions validating our ex vivo results. Further single-cell analyses of the mixed tumor/ lung organoid model resulted in the identification of candidate molecular pathways responsible for the initiation of the observed fibrotic response to tumor cell contact, which we validated using pathway-specific pharmacological inhibitors. Further, in a collaboration with the Medical University Vienna, we studied the consequences of environmental microplastic exposure in our complex lung organoid model and identified pronounced alterations in the transcriptome of healthy lung epithelial cells. Together, our results show that it is possible to model distinct pathological processes in the lung using lung organoids thus providing a solid foundation for future therapeutic drug testing ex vivo.