Neurological diseases are the leading cause of disability worldwide and the second leading cause of death, with more than 600 disorders affecting the peripheral and central nervous systems. Despite this enormous burden, no curative treatment exists for most of them. The drugs currently available for conditions such as Alzheimer’s disease, Parkinson’s disease, or glioblastoma can only ameliorate symptoms; they do not reverse neurodegeneration. Is there a problem upstream? Are we using the right models to develop and test those drugs?
Animal models and classical 2D cell cultures fail to faithfully replicate the complexity of the human brain, leaving a persistent gap between preclinical testing and clinical reality. Brain organoids offer a promising alternative to bridge the gap, yet their full potential in drug discovery remains unrealized.
Why Consistency in Cerebral Organoid Research is a Growing Concern in Drug Discovery
The path from a promising drug candidate to a clinical trial is long, expensive, and, too often, unsuccessful. Approximately 90% of drugs fail in clinical trials, with 70-80% of failures attributed to insufficient efficacy or unacceptable toxicity, problems that preclinical models failed to predict. Brain organoids have been proposed as a more predictive alternative, capable of modeling patient-specific genetic backgrounds and tissue-specific responses.
Studies have already demonstrated their utility:
- Alzheimer’s disease: Park et al. (2021) describe the generation of cerebral organoids from human pluripotent stem cells from sporadic Alzheimer’s patients and used them to screen a selection of FDA-approved drugs. All candidate compounds tested, alone or in combination, proved effective in reducing amyloid-β or tau deposition while preserving neuronal viability, validating the platform as a tool for patient-specific drug testing.
- Glioblastoma: Zhang et al. (2021) implanted patient-derived glioma cells into human cerebral organoids and tested their response to temozolomide (TMZ). Unlike 2D cultures, the organoid system captured tumor microenvironment effects and TMZ resistance patterns that were consistent with the clinical outcomes of individual patients, demonstrating its value for personalized oncology.
- Creutzfeldt-Jakob disease (CJD): Groveman et al. (2021) infected human cerebral organoids with sporadic CJD brain homogenates and evaluated pentosan polysulfate (PPS) both as a prophylactic and a therapeutic agent. PPS reduced prion-seeding activity and prevented PrP aggregate deposition, showing that organoids can model rare neurodegenerative diseases for which no effective treatment currently exists.
However, for organoid-based assays to generate data that regulatory agencies and pharmaceutical companies can trust, they need to meet a fundamental requirement: reproducibility. Without it, differences in drug response across experiments may not reflect genuine biological effects, making it impossible to draw statistically meaningful conclusions or compare results between laboratories. As the field moves toward regulatory frameworks, including the FDA Modernization Act 3.0, the demand for standardized, traceable, and reproducible organoid workflows has never been greater.
Key Sources of Variability in Cerebral Organoid Generation and Maintenance
Variability in cerebral organoids is not a single problem; it is the cumulative result of several interconnected sources that act at different stages of the workflow.
The starting cell population is the first and perhaps most fundamental source of variation. Different iPSC lines carry distinct genetic and epigenetic backgrounds, which translate into differences in differentiation capacity and organoid quality. Even within a single line, cells cultured under different conditions, feeder-dependent versus feeder-free, can yield organoids with substantially different architectural features and cellular compositions. The presence of acquired mutations, including cancer-related variants, can further alter differentiation outcomes and compromise data fidelity. Monitoring genomic integrity and using multiple lines are, therefore, essential steps toward brain organoid reproducibility.
The extracellular matrix (ECM) represents another major bottleneck. Most organoid protocols rely on Matrigel, a protein mixture derived from mouse sarcoma cells. Its molecular composition is undefined, its mechanical properties vary between batches, and it contains growth factors and xenogeneic contaminants that introduce unpredictable biological effects. The human brain ECM has a markedly different composition from Matrigel, being rich in hyaluronan and chondroitin sulfate proteoglycans, with comparatively little collagen or laminin; thus, Matrigel may provide only limited microenvironmental cues relevant to brain development. This batch-to-batch inconsistency in matrix properties has direct consequences on organoid size, morphology, and functional readouts.
Beyond the matrix, uncontrolled spatial organization during organoid formation leads to structural heterogeneity. In standard protocols, neural progenitor cells self-organize into multiple independent rosettes within a single organoid, yielding structures with variable internal architecture. Even minor fluctuations in initial cell positioning, local nutrient availability, or mechanical boundaries can result in organoids that differ substantially in size, cell composition, and maturation stage, even within the same batch.
Regarding spatial organization, the position of organoids within culture wells can also influence their growth and maturation, partly due to differences in evaporation rates and microenvironmental conditions across the well or even across different wells in the same wellplate.
Finally, manual handling is a critical and often underestimated source of variability. Organoid culture is a labor-intensive process that involves numerous repeated operations (cell seeding, matrix embedding, medium exchange, compound addition, and sample collection), each of which introduces operator-dependent inconsistencies. Beyond the risk of pipetting errors or accidental aspiration during medium changes, manual workflows impose practical limits on throughput and make it difficult to ensure that every organoid in a plate is treated with the same timing, volume, and precision. These constraints are particularly problematic in drug screening contexts, where the parallel processing of large organoid numbers under identical conditions is essential.
How Automation Improves Reproducibility in Cerebral Organoid Workflows
Automation transforms labor-intensive organoid workflows into reproducible, scalable platforms. Automated systems can integrate and standardize critical steps across the entire workflow, including cell seeding, medium exchange, compound dosing, imaging, and data collection, reducing human error while preserving organoid integrity. The integration of imaging modules within the same platform enables continuous morphological monitoring throughout the culture period, without the need to handle plates between steps. When every stage is executed under the same controlled conditions, batch-to-batch variability decreases, and the resulting data becomes statistically comparable across experiments.
A fundamental prerequisite for reproducibility is that organoids remain structurally intact and biologically viable throughout the culture period. If routine operations such as medium exchange physically damage organoids or compromise their viability, any downstream comparison between conditions becomes unreliable.
Two independent studies using the MO:BOT illustrate how automation addresses this. In a collaboration between MO:RE and Organotherapeutics, automating three medium exchanges of midbrain organoids derived from iPSCs over 8 days increased vell viability compared to manual handling (Figure 1a). After 8 days, automated organoids maintained morphological parameters including area, circularity, solidity and roundness (Figure 1b).
Figure 1. a) Midbrain organoids’ viability after 3 media exchanges, either manually or using the MO:BOT. b) Morphological parameters (area, circularity, solidity, and roundness) at day 1, 3 and 8 after automated or manual medium exchanges.
In a second study with the Fraunhofer ITMP in Hamburg, the MO:BOT performed accurate and precise medium exchanges during the maturation stage of cortical organoids derived from iPSCs between day 43 and day 50 of culture. Likewise, automated organoids showed a higher viability (Figure 2a) and maintained organoid area and roundness compared to manual handling (Figure 2b), with no organoids aspirated throughout the process.
Figure 2. a) Cortical organoids’ viability after 4 media exchanges, either manually or using the MO:BOT. b) Morphological parameters (area and roundness) of 60 cortical organoids from one batch cultured manually (yellow dots) and 60 organoids cultured automatically (purple dots) at days 42 and 50 (after 4 media changes).
These results show that automation delivers the precision and culture quality that manual handling cannot consistently guarantee, and it is precisely this consistency, applied uniformly across all wells and all time points, that lays the foundation for reproducible organoid experiments.
Designing High-throughput Assays With Standardized Cerebral Organoids
Achieving reproducibility is a prerequisite for high-throughput screening, but it is not sufficient on its own. The design of assays that can generate statistically robust data across thousands of organoids also requires solving the challenge of scalable production.
Microwell array platforms have emerged as one of the most effective tools for this purpose. By partitioning cells into hundreds or thousands of defined microwells within a single plate, they enable the parallel formation of large numbers of organoids with controlled initial geometry. Narazaki et al. screened 298 FDA-approved compounds across more than 2,400 cortical organoids in a single experiment using a microwell-based platform, a scale that would be impossible to achieve with conventional culture methods.
The integration of standardized protocols with these platforms is equally critical. Pre-verified, ready-to-use differentiation methods, validated for specific organoid types and directly integrated into automated systems, as the ones included in the MO:BOT, remove the need for laboratory-specific protocol optimization and reduce one of the most common sources of inter-laboratory variability. When combined with quantitative quality control parameters, organoid size, circularity, and viability, these workflows enable the selection of uniform, biologically consistent organoids before compound exposure begins, ensuring that observed drug responses reflect genuine biological effects rather than differences in organoid quality.
Together, automated cerebral organoid culture, scalable production platforms, and standardized protocols are transforming organoid research from a highly variable, low-throughput discipline into a reliable, high-fidelity tool for drug discovery. Closing the gap between the promise of brain organoids and their routine application in preclinical research will depend precisely on the ability to generate them consistently, at scale, and under conditions that regulatory agencies can trust.
Sources
Giorgi C, Lombardozzi G, Ammannito F, Scenna MS, Maceroni E, Quintiliani M, d’Angelo M, Cimini A, Castelli V. Brain Organoids: A Game-Changer for Drug Testing. Pharmaceutics. 2024 Mar 22;16(4):443. doi: 10.3390/pharmaceutics16040443.
Groveman BR, Ferreira NC, Foliaki ST, Walters RO, Winkler CW, Race B, Hughson AG, Zanusso G, Haigh CL. Human cerebral organoids as a therapeutic drug screening model for Creutzfeldt-Jakob disease. Sci Rep. 2021 Mar 9;11(1):5165. doi: 10.1038/s41598-021-84689-6.
Kim D, Youn J, Kim J, Lee J, Yoon J, Kim DS. From organoid culture to manufacturing: technologies for reproducible and scalable organoid production. npj Biomedical Innovations. 2026; 3:12. doi: 10.1038/s44385-025-00054-6
Pagliaro A, Artegiani B, Hendriks D. Emerging approaches to enhance human brain organoid physiology. Trends Cell Biol. 2025 Jun;35(6):483-499. doi: 10.1016/j.tcb.2024.12.001.
Park JC, Jang SY, Lee D, Lee J, Kang U, Chang H, Kim HJ, Han SH, Seo J, Choi M, Lee DY, Byun MS, Yi D, Cho KH, Mook-Jung I. A logical network-based drug-screening platform for Alzheimer’s disease representing pathological features of human brain organoids. Nat Commun. 2021 Jan 12;12(1):280. doi: 10.1038/s41467-020-20440-5.
MO:RE GmbH. Automated Culture of Midbrain Organoids derived from iPSCs. Application Note, in collaboration with Organotherapeutics. September 2024.
MO:RE GmbH. Automated Medium Exchange of iPSC-derived Cortical Organoids. Application Note, in collaboration with Fraunhofer ITMP Hamburg. February 2024.
Narazaki G, Miura Y, Pavlov SD, Thete MV, Roth JG, Avar M, Shin S, Kim JI, Hudacova Z, Heilshorn SC, Pașca SP. Scalable production of human cortical organoids using a biocompatible polymer. Nat Biomed Eng. 2025 Dec;9(12):2115-2123. doi: 10.1038/s41551-025-01427-3.
Zhang L, Liu F, Weygant N, Zhang J, Hu P, Qin Z, Yang J, Cheng Q, Fan F, Zeng Y, Tang Y, Li Y, Tang A, He F, Peng J, Liao W, Hu Z, Li M, Liu Z. A novel integrated system using patient-derived glioma cerebral organoids and xenografts for disease modeling and drug screening. Cancer Lett. 2021 Mar 1;500:87-97. doi: 10.1016/j.canlet.2020.12.013.
