LAUSR

To the Editor — Human organoids hold tremendous potential for biomedical applications1–3 (Fig. 1a,b). These three-dimensional structures of cultured cells recapitulate important aspects of in vivo organ development and biological function. They provide tractable in vitro models of human physiology and pathology, thereby enabling interventional studies that are difficult or impossible to conduct in human subjects. For example, organoids allow genetic and pharmacological manipulation in a complex cellular context that reflects human biology, and they enable investigations of the early stages of organ development and disease onset. Human organoids complement (and may eventually replace) animal models in many areas of preclinical drug development. Moreover, they provide patient-specific ‘avatars’ for drug development and precision therapies, including treatments for cancer, rare genetic diseases (such as cystic fibrosis) and complex multifactorial disorders (such as epilepsy). Finally, they promise to contribute to regenerative medicine, with the goal of producing functional biological structures that can be transplanted into patients. To realize the full potential of human organoids, key challenges need to be addressed (Fig. 1c). Most immediately, we need better characterization and validation of organoids as faithful models of human biology. This will require assays for informative high-throughput profiling as well as the definition of quality standards for cell composition, cellular differentiation, cell states and responses to stimuli. A future catalog of well-characterized human organoids should include extensive replication, to quantify technical and biological sources of variation. Moreover, inclusion of genetically diverse sample donors will help to assess interindividual variation in the human population. Current organoid protocols, while useful for many applications, have relevant technical and conceptual limitations. For example, organoids may not faithfully represent the diversity of cell types in primary tissue (including non-parenchymal cells such as immune cells and stroma), and they are limited in their ability to account for the effects that environmental exposures and organismal aging have on human organs in vivo. It will be important to develop robust protocols that yield organoids with adequate tissue organization, differentiated cells, vascularization, immune cell infiltration and, for some organs (for example, skin and intestine), even a microbiome. Moreover, we are still learning how to use organoids most effectively for discovering biology (for example, through genetic and pharmacological perturbations) and how to exploit them for drug development and personalized medicine. Single-cell sequencing and spatial profiling have a key role in addressing these issues (Fig. 1d). Comprehensive molecular maps of organoids and organoid development can reveal cell states and transcription regulatory programs in unprecedented detail, and comparisons to corresponding human tissues in vivo provide powerful new ways of evaluating organoids. Single-cell epigenome and transcriptome profiling yields a quantitative, high-dimensional assessment of cell composition and cell states within organoids. Spatial profiling assays characterize tissue organization and three-dimensional architecture. These methods also enhance organoid quality control (for example, identifying outliers, missing cell types or aberrant gene regulation), and they can provide reference atlases for disease-centric studies. Furthermore, comparative molecular profiling of organoids and matched ex vivo tissue samples can guide the development of new and improved organoid protocols, for example, by identifying missing cell populations or bottlenecks of cellular differentiation in organoids. Finally, single-cell technologies provide a powerful and scalable readout for functional experiments and for genetic or pharmacological perturbations that are tested in human organoids. To demonstrate the feasibility and utility of combining human organoids with single-cell technology, we have launched an Organoid Cell Atlas pilot project, as a ‘Biological Network’ within the Human Cell Atlas (HCA)4–6, focusing on the single-cell characterization of organoids and complex in vitro systems (https:// www.humancellatlas.org/coordinators). The Organoid Cell Atlas will facilitate the production, quality control, dissemination and utilization of single-cell and spatial genomics data for human organoids, and it will link such datasets to the comprehensive profiles of primary tissues that are being generated within the HCA. A first step toward establishing the Organoid Cell Atlas has recently been funded by the European Union Horizon 2020 call for “Pilot actions to build the foundations of a human cell atlas” with the ‘HCA|Organoid’ project (https://hca-organoid.eu).