How organoids improve preclinical accuracy

Nature (5 Aug 26) published three coordinated papers that together generate and validate the largest clinically annotated collection of patient-derived 3D cancer models assembled to date, and demonstrate that these models reveal therapeutic gene dependencies undetectable in traditional 2D cell lines. 🔅 The Human Cancer Models Initiative (HCMI) generated 665 models from 637 patients across 25 cancer types, from a consent cohort of 2,780 donors, including 153 rare cancer models and 71 models from non-European donors. Across 421 matched tumour-model pairs, DNA concordance was 97.8% and epigenetic concordance 95%. Single-nucleus RNA sequencing identified three mechanisms of model-tumour divergence: stromal purification, clonal selection, and culture-medium-driven epigenomic plasticity, the last reversible by switching growth conditions. Extrachromosomal DNA was model-specific in 43.9% of cases, yet MYCN, KRAS and EGFR amplifications were preserved in a subset. GBM models from patients with prolonged temozolomide exposure retained the SBS11 treatment resistance mutational signature. All models are available via ATCC, the NCI GDC portal and the HCMI Explorer Suite. 🔅 The Wellcome Sanger Institute derived 256 clinically annotated organoids from five cancer types and completed genome-wide CRISPR-Cas9 screens across 162 organoids (AUROC = 0.97), identifying 97 core fitness genes unique to 3D organoid cultures, enriched in isoprenoid and steroid biosynthesis. In colorectal organoids, KRAS dependency varied by allele: G12X variants retained co-dependency on upstream EGFR signalling while Q61H organoids were fully unresponsive to EGFR inhibition, consistent with signalling-independent oncogenic activation. In paired pre- and post-treatment oesophageal organoids, chemotherapy-driven clonal evolution reduced KRAS and DNMT1 dependency and increased PSMB5 dependency, confirmed pharmacologically with proteasome inhibitors bortezomib and ixazomib. 🔅 The Broad Institute added 314 NextGen models across 10 cancer types to DepMap via 147 CRISPR screens. NextGen models matched annotated cancer lineage in 69% of cases versus 35% for traditional cell lines; CNS spheroids matched at 93% versus 11%. A PDAC-classical/mucinous transcriptional program, preserved in GI organoids but silenced in 2D lines, correlated with WNT pathway dependency (WLS, MESD, FZD5, LGR4 and TCF7L2). In glial GBM spheroids, CDKN2A loss predicted CDK6 dependency and CDK4/6 inhibitor sensitivity. Direct comparison of 3D versus 2D conditions showed that serum-containing media masked SCD dependency in KRAS-amplified oesophagus-stomach organoids, demonstrating that growth medium independently shapes gene essentiality. 📑 HCMI compendium: https://lnkd.in/gjgFTp-3 📑 Sanger organoid biobank: https://lnkd.in/gFXhqeii 📑 Broad DepMap NextGen: https://lnkd.in/ggbssEH4 #CancerResearch #PrecisionOncology #CRISPR

Stanford University researchers have achieved a major milestone in regenerative medicine by creating lab-grown heart and liver organoids that include their own blood vessels.  These vascularized organoids mark a significant leap forward because previous versions lacked the ability to grow beyond a limited size due to the absence of internal circulation.  Without vessels, cells in the center of organoids would die from a lack of nutrients and oxygen. The addition of functional blood vessel networks allows the mini-organs to sustain more realistic structure, function, and longevity in lab environments. The team focused primarily on developing heart organoids that resemble a human embryonic heart around 6.5 weeks into gestation.  These lab-grown hearts were able to beat and exhibited cell diversity that mirrors actual early-stage human heart development.  Using a rigorous screening of 34 different combinations of growth signals and structural scaffolding, researchers identified the optimal formula (referred to as “condition 32”) that reliably produced key cell types including cardiomyocytes (heart muscle cells), smooth muscle, and endothelial cells (which form blood vessels). One striking application was testing how fetal tissues might respond to external chemicals.  The team exposed the heart organoids to fentanyl — a powerful opioid — and observed abnormal increases in vessel growth, indicating that the organoids can model how drugs affect early development.  This has major implications for studying toxic exposures and congenital disease origins. Moreover, the researchers replicated their vascularization technique in liver organoids, proving that the approach is flexible and scalable.  The long-term goal is to create implantable organoids for use in human therapy, especially for repairing damaged organs.  Until then, these mini-organs provide a valuable new platform for drug testing, developmental biology, and reducing animal use in research — aligning with updated FDA guidelines favoring human-relevant models. https://lnkd.in/gRYxbu8R #RegenerativeMedicine #Organoids #VascularizedOrganoids #TissueEngineering #StemCellResearch #BiomedicalInnovation #HeartOrganoids #LiverOrganoids #Cardiomyocytes #VascularBiology #OrganRegeneration #MiniOrgans #LabGrownOrgans #DrugTesting #DevelopmentalBiology #ToxicologyResearch #FetalHealth #HumanRelevantModels #AlternativesToAnimalTesting #FutureOfMedicine #FDAInnovation #NextGenTherapies #PrecisionHealth #MedicalBreakthrough

I woke up to this news that: Scientists Just Solved Organoids' Biggest Problem! I’m happy to share highlights from a new Science paper by Dr. Oscar Abilez, Dr. Huaxiao 'Adam' Yang, Dr. Joseph C. Wu, and colleagues, a leap forward for organoid technology and regenerative medicine! What Did They Do? Stanford researchers have created the first heart and liver organoids with integrated, functional blood vessels. This solves a critical bottleneck: until now, organoids could only grow a few millimeters before their centers died from lack of oxygen and nutrients. With built-in vasculature, these mini-organs can grow larger, mature further, and better mimic real human tissues. How Did They Do It? *The team meticulously optimized a “recipe” of growth factors and signaling molecules, guiding pluripotent stem cells to differentiate into not just heart or liver cells, but also endothelial and smooth muscle cells that self-organize into branching blood vessels. *Their protocol mirrors early embryonic development, allowing the organoids to achieve a cellular complexity similar to a 6.5-week-old human embryonic heart, including beating function! Why Is This Important? *Better Disease Models: Vascularized organoids allow researchers to study early human development and test how drugs impact organ growth and blood vessel formation. *Personalized Medicine: These models can be tailored from patient-derived stem cells, paving the way for individualized drug testing and disease modeling. *Regenerative Therapies: In the future, vascularized cardiac organoids could be implanted to repair damaged heart tissue, offering a more complete cellular environment than current cell therapies Clinical Context As Dr Joseph C. Wu notes, ongoing clinical studies are already injecting lab-grown cardiomyocytes into patients with heart dysfunction. But real heart tissue is much more complex, containing blood vessels, pericytes, fibroblasts, and more. Vascularized organoids could one day provide all these cell types in a single, implantable tissue patch, dramatically improving integration and function. What’s Next? The team aims to: *Grow organoids longer to assess their maturation and size limits *Further refine the recipes to include immune and blood cells *Adapt this vascularization approach to other organs, moving closer to true “mini-organs” for research and therapy A huge CONGRATULATIONS to the entire Stanford team! References: https://lnkd.in/gmYc-cX9 https://lnkd.in/gbntyWgN https://lnkd.in/g-YT5wdU

Our latest paper is out in Nature today! Zonation is a design principle of the human liver. For decades, scientists have known that liver function varies by zone — with diseases like DILI (Drug-Induced Liver Injury) and MASLD often striking specific areas near the central vein. Yet, no human stem cell-derived model has been able to reproduce this complexity — until now. In this study, we report the first human iPSC-derived liver organoids that exhibit controlled zonation, capturing real metabolic gradients across the lobule. 👉 Read the full paper: https://rdcu.be/eh3nX • We restored a functional Gulo gene (defective in humans) to induce an endogenous ascorbate gradient, layered with bilirubin to simulate natural zonation cues. • Zonally primed organoid can fuse to self-assemble and reconstruct all three zones. • Zone-specific functions and toxicities revealed: several drugs target only zone 1 or zone 3 hepatocytes. • Elastic epigenetic regulation uncovered — driven by TET1 (ascorbate) and HIF1A (bilirubin). • Multi-zonal organoid transplantation reduces total liver dysfunction as a promising strategy for regeneration. This project was led by my first US based PhD student, Hasan Al Reza, PhD, who immigrated from Bangladesh to pursue science with passion and resilience. Deep gratitude to the team at Cincinnati Children's for pushing the boundary of human organoid models. #Organoids #LiverZonation #StemCells #GuloGene #AscorbateGradient #RegenerativeMedicine #iPSC #DrugDiscovery #PrecisionMedicine #DiversityInSTEM #NaturePaper #DILI

Brain organoids just modeled a pain circuit from sensory receptor to cortex. Four fused regional organoids. Capsaicin applied at the sensory end. Electrical response measured at the cortical end. Let that satisfying experimental logic sink in for a moment. Now here is the part that matters for anyone in the preclinical translation space. This is not a replacement for in vivo pain research. It is a new resolution layer. One that lets you isolate circuit-level responses without the confound of systemic physiology. And one that lets you test human-specific biology that no rodent model can provide. The Nature feature published today walks through a decade of progress. From Lancaster's first accidental clumps in 2010 to Pașca's assembloids that now connect cortex, spinal cord, and muscle tissue in a dish. Organoids that contract muscle when you stimulate spinal neurons. Organoids that model poliovirus-induced paralysis. Organoids built from Neanderthal gene variants to study evolutionary divergence in neural maturation. The science is real. The translational potential is real. But so are the limitations. Organoids survive months, not years. They lack vasculature. They lack immune context. They lack the integrated systemic feedback that defines whole-organism biology. This is exactly why the framing matters. The question is not: can organoids replace animal models? The question is: what specific biological questions can organoids answer that no other system can? Human-specific progenitor cell dynamics. Species-specific maturation timing. Circuit-level pharmacology in human neural tissue. These are contexts of use. Not ideology. Not replacement narratives. Not wishful extrapolation. The field is at what Knoblich calls an inflection point. I agree. But inflection points are where precision in framing matters most. Because the fastest way to discredit a powerful technology is to oversell what it does. Organoids are not miniature brains. They are miniature questions, asked in human tissue, at a resolution we never had before. That is more than enough. #Organoids #NAMs #PreclinicalResearch #Neuroscience #DigitalBiomarkers #TranslationalScience #BrainResearch

Can brain organoids mimic the rhythms of a developing human brain? Trujillo et al. recorded electrical activity from cortical organoids over time. By month 6, these 3D models began to show network bursts: brief, synchronized waves of activity. This figure compares these bursts to EEG signals from preterm infants. The resemblance is striking: both display spontaneous activity transients (SATs).. short bursts punctuating otherwise quiet periods, a hallmark of early human brain development. This suggests that human brain organoids don’t just look like fetal brains under the microscope, but they might also function like them, at least in terms of network-level activity. This was a powerful step toward bridging in vitro models and translational neuroscience. 🔗 Link to the paper in the comments. 📍 Follow for more neuroscience, health and technology #Neuroscience #Organoids #Electrophysiology #EEG #BrainDevelopment #StemCells #Neurotech

Dual CAR-T: Brain tumor organoids accurately model patient response to CAR T cell therapy - Penn Medicine, University of Pennsylvania Health System PHILADELPHIA— For the first time, researchers used lab-grown organoids created from tumors of individuals with glioblastoma (GBM) to accurately model a patient’s response to CAR T cell therapy in real time. The organoid’s response to therapy mirrored the response of the actual tumor in the patient’s brain. That is, if the tumor-derived organoid shrunk after treatment, so did the patient’s actual tumor, according to new research from the Perelman School of Medicine at the University of Pennsylvania, published today in Cell Stem Cell. “It’s hard to measure how a patient with GBM responds to treatment because we can’t regularly biopsy the brain, and it can be difficult to discern tumor growth from treatment-related inflammation on MRI imaging,” said Hongjun Song, PhD, the Perelman Professor of Neuroscience and co-senior author of the research. “These organoids reflect what is happening in an individual’s brain with great accuracy, and we hope that they can be used in the future to ‘get to know’ each patient’s distinctly complicated tumor and quickly determine which therapies would be most effective for them for personalized medicine.” Patient-derived glioblastoma organoids as real-time avatars for assessing responses to clinical CAR-T cell therapy Highlights •Rapid GBO generation in a phase 1 clinical trial for real-time treatment assessment •A unique trial design with parallel treatments of patients and patient-derived GBOs •Cell cytolysis and target antigen reduction in GBOs treated with patient CAR-T cells •Cytokine release and the degree of cytolysis in GBOs correlated with patient metrics Summary Patient-derived tumor organoids have been leveraged for disease modeling and preclinical studies but rarely applied in real time to aid with interpretation of patient treatment responses in clinics. We recently demonstrated early efficacy signals in a first-in-human, phase 1 study of dual-targeting chimeric antigen receptor (CAR)-T cells (EGFR-IL13Rα2 CAR-T cells) in patients with recurrent glioblastoma. Here, we analyzed six sets of patient-derived glioblastoma organoids (GBOs) treated concurrently with the same autologous CAR-T cell products as patients in our phase 1 study. We found that CAR-T cell treatment led to target antigen reduction and cytolysis of tumor cells in GBOs, the degree of which correlated with CAR-T cell engraftment detected in patients’ cerebrospinal fluid (CSF). Furthermore, cytokine release patterns in GBOs mirrored those in patient CSF samples over time. https://lnkd.in/eusyxeba https://lnkd.in/es-s4MSt

Retinal organoids are becoming functionally mature enough to improve drug testing for eye diseases. A new study led by Volker Busskamp at University Hospital Bonn combines human stem cell-derived retinal organoids with endothelial cells, enabling the formation of vascular-like networks that deliver oxygen and nutrients. This is huge because retinal ganglion cells, which transmit visual information from the eye to the brain, are difficult to maintain in organoid systems. → The organoids reacted to light, and signals were correctly transmitted to the ganglion cells. → Organoids also responded to hypoxia. Under hypoxic conditions, the artificial vessels remodeled and expanded, mirroring changes observed in retinal disease. This moves retinal organoids closer to modeling conditions such as retinopathy and enables drug testing in a more human-relevant in vitro system. There is, however, a tradeoff. As organoids become more biomimetic, they also become harder to handle. Complex models will require more robust systems, and we're working to help on that. 🔗 Check the paper: https://lnkd.in/e3Q8jyiu 🔬 Vascularised retinal organoids. Credit: University Hospital Bonn.

I'm happy to share our latest review article on "Integrating Human Intestinal Organoids into FDA's New Approach Methodologies for Drug Discovery" published in in Advanced Science. More than 85% of IBD drug candidates fail in clinical trials. A key reason is that animal models fail to predict human safety early, a translational gap that has cost the field decades and billions of dollars. The FDA's Modernization Act 2.0 (2022) and NAMs guidance (2025) are now formally opening regulatory submissions to human-relevant, non-animal data, organoids, organ-on-chip, microphysiological systems, and AI-integrated platforms. But replacing animals in drug testing is not straightforward. The FDA's expectation is precise: the right tool, properly validated for the specific regulatory question. This is the foundation of Context of Use and Fit-for-Purpose validation and it sets a high bar. In this review, we assess where human intestinal organoids (HIOs) meet that bar and where work remains. On the promising side, HIOs express relevant drug transporters and metabolic enzymes, preserve patient-specific genetic and epigenetic features, and have successfully identified clinical GI toxins in toxicity assays. Over 30 registered clinical trials for intestinal diseases now deploy patient-derived organoids, and organoid-guided decisions have already reached patients in cystic fibrosis through theratyping. However, significant challenges remain. Current HIOs largely lack immune cells, vasculature, and a functional microbiome component that are central to IBD pathology and drug response. Reproducibility across laboratories remains inconsistent, partly due to variability in culture conditions and extracellular matrix batches. Closed organoid lumens limit direct apical drug exposure. And standalone intestinal organoids cannot independently model systemic pharmacokinetics requiring integration with liver, kidney, and other major organs through microphysiological systems, which adds substantial complexity. The key takeaway is that intestinal organoids hold genuine regulatory potential, but realizing it requires context-specific validation, standardized protocols, and transparent reporting of where these models perform well and where they do not. Our ongoing IBD drug discovery research is actively integrating NAM-based approaches, with a clear-eyed view of both their current utility and their limitations. Many thanks to authors Debarun Patra, Ibrahim M Sayed, Souhrid Mukherjee, Honit Piplani, Aida Habtezion, and Michael Rosen, from the Stanford University School of Medicine, Stanford Medicine Children's Health, Stanford Cardiovascular Institute, Greenstone Biosciences, and Cibus Therapeutics. Full text: https://lnkd.in/gWAxT_CU

Can organoids model human brain development? Two papers from the last couple of days dive deep into this question. The first, a review from Miguel Forte Tenreiro and Alysson Muotri, describes the extent to which cerebral organoids recapitulate corticogenesis. The progenitor hierarchy is present (neuroepithelial cells, apical radial glia, outer radial glia and intermediate progenitors), the broad inside-out sequence of neurogenesis is seen, as is the emergence of oscillatory network activity over time. However, laminar organisation is rudimentary, the outer subventricular zone (oSVZ) is underdeveloped, and upper layer neuron subtypes are underrepresented. The review is clear about these gaps, but also emphasizes that cerebral organoids remain a valuable tool, particularly for early neurodevelopmental disease modelling. The second paper, from Shreyash Sonthalia, Carlo Colantuoni and collaborators, goes beyond the review to describe new research that takes a quantitative approach to the same question. Using joint transcriptomic decomposition across mouse, macaque and human datasets, they define layer-specific neuronal maturity signatures from the adult cortex and then examine whether organoids map them. The answer is partially; organoids show deep layer transcriptomic signatures, specifically subplate and L5/6, but upper layer programs, L2/3 and the primate-specific L4 signature, are largely absent. They also show that the transcription factors that specify laminar identity peak early in fetal development and then decline, while the full laminar transcriptome they drive continues building for years postnatally. Organoids recapitulate the former, but not most of the latter. The Tenreiro/Muotri review suggests transplantation into the rodent cortex as an option to extend neuronal maturation. Sonthalia et al. test this directly and find it is beneficial for most deep layer signatures, but not for L2/3 and not at for primate-specific L4 programs. The rodent environment can't induce what hasn’t developed evolutionarily. Neither paper is pessimistic about organoids, and both identify what needs to improve to be able to model later corticodevelopment: better oSVZ expansion, longer maturation windows, and more representative benchmarks for what "layer identity" actually means transcriptomically. Read together, these publications are a fantastic pairing - one describes the capabilities and limitations of brain organoids, and the other quantifies them. https://lnkd.in/gz45-dv6 https://lnkd.in/g8TR7_dU