1. Introduction
The development of safe and effective therapies in tissue engineering and regenerative medicine (TERM) faces a critical bottleneck: the inadequacy of conventional preclinical models. Traditional two-dimensional (2D) cell cultures and animal models fundamentally fail to recapitulate the architectural and functional complexity of human tissues, leading to unreliable predictions of drug efficacy, toxicity, and host integration [
1], [
2], [
3]. These models are particularly limited for evaluating TERM strategies, which often involve biomaterial scaffolds, stem cells, and engineered tissues that operate within a dynamic, multicellular microenvironment. Persistent interspecies discrepancies in drug metabolism and immune response further diminish the translational value of animal data [
4,
5], while ethical concerns regarding animal use continue to increase [
6].
Organoid technology has emerged as a transformative platform to overcome these limitations. These three-dimensional (3D), self-organizing microtissues derived from stem or progenitor cells faithfully mimic the structural and functional properties of human organs [
7,
8]. Their ability to model patient-specific pathophysiology and genetic diversity makes them particularly valuable for TERM applications, enabling ① precise assessment of organ-specific drug toxicity and metabolism, ② development of personalized therapeutic strategies, and ③ evaluation of complex host-implant interactions [
9], [
10], [
11]. Furthermore, organoids demonstrate exceptional utility in modeling multicellular crosstalk involving endothelial, immune, and stromal cells, providing unprecedented insight into drug mechanisms and regenerative processes [
12,
13].
Despite their promise, significant challenges impede the widespread adoption of organoids in drug development pipelines. Issues of reproducibility, functional maturation, vascularization, and scalability must be addressed to realize their full potential [
14,
15]. Patient-derived organoids (PDOs) can be used to evaluate drugs on an individualized basis, allowing for more personalized medicine and tailored therapeutic strategies [
16]. Organoids reduce the need for animal testing, addressing both ethical concerns and regulatory requirements [
17]. By providing a more accurate model for human drug testing, organoids can reduce reliance on animal models, align with ethical guidelines, and improve the public perception of drug development practices. As regulatory agencies increasingly recognize the value of organoids in drug testing, these models may facilitate the approval process by providing more relevant preclinical data [
14,
18]. Organoids are useful not only for assessing adverse effects but also for understanding the mechanisms of drug toxicity. These approaches allow researchers to study how drugs interact at the molecular level within organoids and identify specific genes or proteins involved in drug responses [
19,
20]. Some organoids are engineered with fluorescent reporters or other tracking systems, enabling real-time assessment of drug-induced changes in cellular behaviors, such as hypertrophy, apoptosis, or metabolic shifts [
21].
Organoids play a transformative role in advancing TERM, as they offer unique advantages in addressing unmet clinical needs, particularly in fields such as organ failure, degenerative diseases, and cancer [
22,
23]. These models help bridge critical gaps in current therapies, offering new possibilities for patient-specific treatments, drug development, and tissue regeneration [
24]. Organ failure remains among the most pressing challenges in modern medicine, with limited treatment options and a shortage of suitable donor organs for transplantation [
25]. One of the most transformative aspects of organoid technology in the context of TERM is its ability to enable patient-specific therapeutic strategies. This personalization of treatment has the potential to revolutionize the way we approach regenerative applications, such as liver regeneration and neural repair, by tailoring therapies to an individual’s unique genetic makeup, disease state, and response to treatment [
24,
26,
27].
This review aims to critically evaluate the evolving role of organoid technology in drug evaluation within the specific context of TERM. We will discuss the significant advances enabled by various organoid models—including hepatic, cardiac, neural, gastrointestinal, lung, and tumor organoid models—in assessing organ-specific toxicity, predicting human-specific drug responses, and enabling personalized therapeutic screening. Furthermore, we will explore innovative strategies, such as organ-on-a-chip (OOC) integration and 3D bioprinting, which enhance the physiological relevance and scalability of these models. A central focus of this review is to honestly address the persistent challenges that hinder their widespread adoption, including issues of reproducibility, functional maturity, limited vascularization, and immune integration. Finally, we will navigate the critical pathway toward clinical translation by examining the ethical considerations, regulatory landscape, and standardization efforts required to bridge the gap between pioneering research and routine clinical application. By providing a comprehensive analysis of both the potential and pitfalls, this review seeks to chart a course for the future development and integration of organoid-based assays to create safer and more effective regenerative medicines (
Fig. 1).
2. Classification and coverage of organoid models
2.1. Classification and application of organoid models by physiological system
To systematically evaluate the role of organoids in TERM drug development, it is essential to categorize them on the basis of the physiological system they represent [
28]. This approach provides a clear rationale for model selection, highlighting the unique functional aspects of each organ type that are critical for assessing drug efficacy, toxicity, and regenerative potential. The most advanced and widely used organoid models in TERM can be classified into the following systems: metabolic, excitable, and barrier/lining tissues, with tumor organoids representing a unique cross-cutting category (
Table 1 [
15], [
29], [
30], [
31], [
32], [
33], [
34], [
35], [
36], [
37], [
38], [
39], [
40], [
41], [
42], [
43], [
44], [
45], [
46], [
47], [
48], [
49]).
2.1.1. Organoids in the metabolic system
As the primary site of drug metabolism and detoxification, the liver is a critical organ for toxicity assessment. Liver organoids bridge this gap by replicating key architectural and functional aspects of the liver. They exhibit metabolic functions, including cytochrome P450 (CYP450) enzyme activity, enabling the study of drug-induced liver injury (DILI), chronic toxicity, and drug-drug interactions. The ability to derive these organoids from patient-specific induced pluripotent stem cells (iPSCs) or primary tissues allows for personalized hepatotoxicity testing and the modeling of genetic liver diseases, making them indispensable for predicting human-specific metabolic responses [
29], [
30], [
31], [
50].
2.1.2. Excitable tissue organoids
Cardiac organoids model the electromechanical properties of the human heart. They are crucial for assessing drug-induced cardiotoxicity, a major cause of drug attrition, particularly in oncology. These models recapitulate electrical activity and contractility and can detect arrhythmogenic effects more accurately than traditional 2D models can. Patient-derived cardiac organoids facilitate personalized risk assessment for cardiovascular complications, enabling the development of safer therapeutic regimens [
32,
33].
Neural organoids have advanced beyond generic models to include region-specific constructs that mimic distinct areas of the human brain. PDOs can model genetic predispositions [
37]. They are used to investigate neuroinflammatory mechanisms (microglial response, cytokine release, and signaling pathways), oxidative stress, excitotoxicity, cell death, and synaptic dysfunction [
51]. Personalized neurotoxicity testing and screening for neuroprotective drugs are key applications, potentially improving preclinical safety assessments and aiding in the treatment of neurodegenerative disease [
38]. This specificity is critical for accurately modeling diseases and assessing neurotoxicity, which often affects particular brain regions [
34], [
35], [
36], [
52].
(1) Cerebral cortical organoids (or forebrain organoids) model the cerebral cortex and are extensively used to study neurodevelopmental disorders, neurodegenerative diseases, and the effects of drugs on improved cognitive function. They allow for the study of drug-induced effects on cortical neuronal networks, neuroinflammation, and synaptic dysfunction [
34,
35].
(2) Midbrain organoids specifically recapitulate the dopaminergic pathways of the midbrain. They are indispensable for modeling Parkinson’s disease, screening neuroprotective agents for dopaminergic neurons, and assessing the neurotoxic potential of compounds known to target this region [
34,
36].
(3) Hindbrain organoids can model structures such as the medulla oblongata and cerebellum, which is relevant for studying certain neurodevelopmental issues and metabolic regulation. The incorporation of patient-derived cells into these region-specific models allows for the modeling of genetic predispositions to neurological disorders and the personalized screening of neuroprotective agents, thus drastically improving the predictive value of preclinical safety assessments [
37,
38].
2.1.3. Barrier and lining tissue organoids
Intestinal organoids model the complex structure of the gut epithelium, including the crypt-villus architecture and diverse cell types (enterocytes, goblet cells, and Paneth cells) [
39,
40]. They are used to study drug absorption, gut metabolism, and gastrointestinal toxicity [
41,
42]. Furthermore, they can be leveraged to model host-microbiome interactions and inflammatory bowel diseases, providing a relevant system for evaluating oral drug formulations and biologics [
53,
54].
Lung organoids recapitulate the proximal and distal regions of the respiratory epithelium [
43]. They are applied in studies of inhalation toxicology, fibrotic lung diseases, and host responses to respiratory infections [
44]. These models are particularly valuable for evaluating the toxicity of inhaled pharmaceuticals and environmental toxins, as well as for developing therapies for regenerative lung medicine [
45,
46].
2.1.4. Oncology (cross-cutting) organoids
Tumor organoids derived from patient biopsies maintain tumor heterogeneity and microenvironmental features, making them excellent for oncology drug screening [
47]. While derived from specific organs, PDOs constitute a distinct category because of their application in personalized oncology. PDOs retain the genetic heterogeneity and microenvironmental features of the original tumor. They are primarily used for high-throughput drug screening to predict individual patient responses to chemotherapy, targeted therapy, and immunotherapy. This enables the selection of optimal treatment regimens, identification of resistance mechanisms, and development of novel anticancer therapies, representing a pinnacle of personalized medicine in TERM [
15,
47,
48]. PDOs are invaluable for studying rare genetic disorders and offer platforms for disease modeling, drug screening, and personalized treatment. They replicate patient-specific disease characteristics, allowing for the testing of small molecules, gene therapy, and protein replacement therapies [
49]. PDOs enable individualized treatment strategies on the basis of a patient’s unique genetic profile [
16]. Clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated protein 9 (Cas9) editing can create accurate disease models, serving as testbeds for gene editing and stem cell therapies [
55].
2.2. Comparison with traditional models
Traditional 2D cell cultures grown on flat surfaces fail to replicate the 3D structure and functional organization of human tissues [
56]. This severely limits their ability to model complex cell-cell interactions crucial for accurate drug response and toxicity assessments. They also lack the dynamic physiology (nutrient diffusion, waste removal, and biochemical gradients) present in real organs, making them more physiologically relevant, with their 3D environment and self-organization into tissue-like structures [
57] (
Table 2 [
5], [
14], [
15], [
16],[
56], [
57], [
58], [
59]).
Animal models, while historically used, often fail to accurately predict human drug metabolism and toxicity because of species-specific differences in enzymes, receptors, and organ functions, leading to high clinical trial failure rates [
50]. Their immune systems also differ significantly from those of humans, influencing the testing of biologics and immunotherapies. Furthermore, ethical concerns surrounding animal testing are growing. Organoids derived from human cells offer a more human-relevant and ethically sound alternative [
5].
Organoids combine
in vitro and
in vivo strengths while overcoming the limitations of traditional models. Their human origin provides genetic and physiological relevance and more accurately predicts human drug responses than animal models do. Organoids are amenable to high-throughput screening, allowing rapid and large-scale drug efficacy and toxicity testing and accelerating drug development. Miniaturized organoid platforms further enhance automation and reproducibility [
59]. Organoids offer significant ethical advantages by reducing the need for animal testing and enabling personalized medical approaches using patient-specific cells.
3. Methodological innovations enhancing organoid utility
3.1. Advancements in organoid fabrication and analysis for enhanced drug evaluation
The physiological relevance of organoids in drug testing is considerably enhanced by the integration of advanced technologies. These innovations can be categorized into two complementary areas: ① manufacturing and fabrication technologies that enable the construction of more complex, vascularized, and interconnected organoid systems and ② characterization and analytical platforms that provide deep, multiparametric readouts of drug effects, moving beyond simple viability assays (
Table 3 [
59], [
60], [
61], [
62], [
63], [
64], [
65], [
66], [
67], [
68], [
69]).
3D bioprinting enables the precise layer-by-layer deposition of cells, biomaterials, and growth factors to create complex, vascularized organoid structures. It is crucial for improving nutrient and oxygen delivery, which is essential for long-term culture and functional maturity. Bioprinting allows the creation of organoids with defined architectures, such as hepatic cords or cardiac patches, enhancing their physiological accuracy for chronic toxicity studies and regenerative medicine applications [
60,
61]. For instance, bioprinted vascularized liver organoids exhibit improved metabolic activity, whereas cardiac tissues exhibit superior electrical conductivity and contractile behavior for chronic cardiotoxicity assessment [
62,
70]. Current challenges include improving bioprinter resolution and developing novel bioinks with optimal biocompatibility and mechanical properties.
Microfluidic platforms represent a pivotal innovation for organoid culture and interconnection. These systems provide a controlled microenvironment with continuous perfusion, ensuring efficient nutrient delivery and waste removal. Compared with static cultures, these dynamic culture conditions increase organoid growth, reduce necrosis in larger organoids, and significantly improve batch-to-batch reproducibility and functional maturity [
59,
63]. This makes them ideal for scalable and standardized organoid generation.
OOC platforms take this further by housing organoids in microfluidic channels lined with living cells and subjecting them to biomechanical cues. They are used for studying interorgan interactions and systemic toxicity. For example, lung chips mimic breathing motions to reveal variations in aerosolized drug toxicity, whereas gut-liver-kidney chips map organ-specific toxicity following intestinal drug absorption [
64,
65,
71]. Integrating multiple organoids into a single OOC platform allows real-time assessment of metabolic interactions and off-target effects of drug metabolites, providing a holistic view of drug toxicity profiles [
65,
66].
The integration of genomics, transcriptomics, proteomics, and metabolomics with organoid research provides unprecedented insights into drug mechanisms of action at the molecular level. Profiling identifies comprehensive changes in gene expression, protein synthesis, and metabolic pathways in response to drug exposure [
67,
68]. For example, transcriptomic analysis of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-infected lung chips [
72] or metabolomic profiling of liver-islet systems under high glucose conditions can elucidate host-pathogen interactions and disease-specific drug responses [
73]. Blood-brain barrier (BBB) chips simulate nanoparticle transport, whereas lung chips mimic breathing to reveal toxicity variations [
74]. Multiomics approaches are powerful for identifying novel biomarkers of efficacy and toxicity, validating drug targets, predicting idiosyncratic side effects, and informing personalized treatment strategies [
67].
Moving beyond omics, organoids enable a suite of functional analyses. Real-time live-cell imaging can track cellular behaviors such as migration, apoptosis, and metabolic shifts. Electrophysiological analysis using microelectrode arrays (MEAs) is standard for detecting functional changes in network activity or beat rhythm in neural and cardiac organoids. Furthermore, the integration of biosensors allows for the continuous monitoring of metabolites and biomarkers directly from the culture medium, providing nondestructive, kinetic data on organoid function and drug response [
60,
68].
3.2. Current state of drug testing in TERM
While they are the gold standard for preclinical drug testing, they suffer from significant limitations because of metabolic, immune, and regenerative differences between humans and animals [
65]. This leads to poor translation, with many drugs being safe in animal trials because of poor physiology [
14]. Animal models are costly, have long timelines, raise ethical concerns, and poorly simulate cellular-level regeneration, complex 3D tissue architecture, biomechanical forces, and cell‒matrix interactions [
58].
Owing to the simplicity of high-throughput cytotoxicity screening and basic mechanistic studies, 2D cultures fail to model multiorgan crosstalk and are limited in their ability to study long-term effects [
75]. Grown on flat surfaces, they lack the 3D tissue structure, proper cell signaling, and cellular heterogeneity of
in vivo systems, hindering accurate prediction of systemic toxicity, drug metabolism, and regenerative responses.
A significant challenge in TERM drug testing is the inability of traditional models to accurately predict human responses, especially for stem cell, gene, and organ engineering therapies. Animal and 2D models do not fully replicate human-specific cellular and molecular pathways, including immune responses to foreign implants, biomaterials, or genetically modified cells [
76]. The regenerative potential of human tissues and stem cell differentiation cannot be accurately modeled in animals, limiting the prediction of long-term outcomes [
77]. Traditional models also fail to recreate complex interorgan interactions crucial for assessing the systemic effects of regenerative treatments [
65]. Unpredictable outcomes such as teratoma formation [
78], immune rejection [
79], and maladaptive tissue growth [
80] are difficult to assess in these models. Furthermore, traditional methods are ill equipped to predict long-term regenerative outcomes and do not adequately simulate the complex feedback mechanisms involved in tissue integration. The complexity of regenerative therapies also raises ethical and regulatory challenges, driving the need for more predictive, human-relevant models and new regulatory frameworks. Human-derived tissues and organs hold promise for more relevant data but require new ethical guidelines.
4. Applications in drug evaluation and term
4.1. Organoid models for toxicity testing
Organoids, which are more representative of human tissue than 2D cultures, are used to assess cell damage or death caused by substances via cell viability assays and morphological observation [
32]. Organoids allow for more accurate
in vitro studies of DNA damage from mutagenic compounds using assays such as comet and micronucleus tests, better mimicking the
in vivo cellular environment [
81]. Human-derived liver organoids model DILI by replicating metabolism, enzyme activity, and bile secretion, enabling early detection of liver-toxic compounds [
30]. Cardiomyocyte-derived organoids assess drug effects on heart function and cell viability, recapitulating electrical conductivity, contractility, and responses to ischemia or drug exposure [
82]. Liver organoids, which express CYP450 enzymes, mimic the drug metabolism processes of the liver, enabling the prediction of adverse reactions and the identification of toxic metabolites [
31]. They are also used to model drug-drug interactions, which are crucial in polypharmacy, providing a more human-relevant prediction of drug safety than animal models do.
4.2. Personalized medicine and disease modeling
Integrating multiple organoid models into interconnected OOC platforms can simulate dynamic interorgan interactions [
66]. This allows for comprehensive drug safety evaluations, assessing the simultaneous effects on various organs and off-target effects of drug metabolites. Compared with single-organ models, liver-heart-kidney OOC systems enable real-time assessment of drug-induced toxicity and metabolic interactions, providing more accurate predictions of systemic toxicity and drug interactions [
65].
PDOs enable personalized drug testing, aligning with precision medicine. PDOs from tumor biopsies predict responses to cancer therapies, guiding treatment selection [
20]. Similarly, PDOs are used to assess the impact of regenerative therapies on individuals, considering genetic profiles and disease states to improve treatment precision and safety [
15].
4.3. Drug repurposing and high-throughput screening
Repurposing existing US Food and Drug Administration (FDA)-approved drugs using organoid models is a promising strategy [
83]. High-throughput screening of these libraries on cartilage organoids revealed phenoxybenzamine (an α2-adrenergic receptor (α2-AR) inhibitor) as a potential agent for promoting chondrocyte proliferation and cartilage repair in osteoarthritis [
84]. This approach accelerates drug discovery for tissue regeneration by leveraging the known safety profiles of approved drugs, reducing time and cost for clinical translation. This strategy can be applied to other tissues beyond cartilage to identify compounds for accelerating repair and regeneration.
4.4. Integration with regenerative therapies
Organoids derived from various tissues can assess off-target effects by analyzing changes in gene expression, protein synthesis, and cellular function upon drug exposure [
68]. Multiorganoid systems model the simultaneous impact of drugs on multiple organs, simulate systemic toxicity, and evaluate how metabolites from one organ affect others, leading to more accurate and holistic drug safety evaluations. Companies such as Hesperos, Organovo, and Johnson & Johnson Innovation utilize multiorganoid systems for comprehensive toxicity testing, particularly for regenerative therapies.
5. Challenges and limitations in organoid drug evaluation
5.1. Reproducibility and standardization
The promise of the use of organoids for high-throughput screening and regulatory decision-making is critically undermined by persistent challenges in reproducibility and standardization. While often attributed to simple technical variables, the problem is more profound. It stems from the inherent biological plasticity of stem cells and the complex, often poorly defined, morphogen signaling required for self-organization. This results in the generation of organoids that may be qualitatively similar but quantitatively variable in key functional metrics [
85,
86]. This variability poses a serious challenge for multilaboratory studies and threatens the reliability of organoid-based data for clinical translation. Therefore, claims of organoid utility must be critically evaluated alongside evidence of functional consistency and rigorous benchmarking against known
in vivo responses, not merely morphological similarity. Stem cell quality, especially pluripotency, also affects the efficiency and consistency of organoid formation. The composition of the culture medium, including growth factors and the source of the extracellular matrix (ECM), critically influences organoid growth, differentiation, and functionality [
87]. Minor fluctuations in growth factor concentrations or the use of different ECM batches may contribute to significant interbatch variability, manifesting as inconsistencies in organoid morphology, size, and cellular composition.
This variability in batch quality undermines the reproducibility and reliability of experimental results, especially in high-throughput drug screening and multilaboratory studies. Inconsistent organoid responses to drugs generate conflicting data, hindering the translation of research into clinical applications [
22]. To address this problem, efforts have focused on standardizing organoid-related protocols [
86,
88]. Standardizing the source of stem cells is key. There are initiatives to develop protocols for deriving organoids from human iPSCs or other consistent, ethically sourced progenitors. Using well-characterized, commercial iPSC lines can reduce donor-related variability [
89,
90].
Optimizing the differentiation protocols for organoid generation is crucial. This involves precisely timing, controlling the concentration, and combining growth factors to drive stem cell differentiation into specific organ lineages. There is a push to create standardized media formulations for widespread use in labs. Defining the optimal concentrations of nutrients, growth factors, and ECM components ensures consistent organoid growth. Bioreactor systems are being developed to automate and standardize culture conditions such as temperature, oxygen levels, and mechanical forces, reducing human error and improving batch-to-batch consistency. Standardized protocols for cell seeding density, passaging, and culture maintenance help organoids grow steadily and retain functionality. Automated passaging systems and clear seeding density guidelines minimize variability [
91].
For high-throughput drug testing, organoid models are being adapted with standardized drug administration, exposure time, and endpoint assay protocols [
92]. Automated platforms are used to assess drug effects reproducibly, and standardized functional readouts for drug testing enable cross-lab and cross-experimental data comparison [
69]. Standardized assays also consistently detect toxicological effects. In multiorgan systems, standardizing how organoid models from different tissues are connected in microfluidic platforms is essential to replicate physiological interorgan interactions for evaluating drug systemic effects and predicting multiorgan toxicity.
5.2. Inherent limitations in recapitulating whole-organ functions
While organoids are powerful tools for modeling human tissues, conventional models possess significant inherent biological limitations that restrict their accuracy for drug evaluation. These limitations primarily stem from their inability to fully recapitulate the multicellular complexity and systemic integration of native organs. These are ① the absence of functional immune components; ② a lack of vascularization and perfusion; and ③ incomplete cellular diversity and an incomplete microenvironment.
Standard organoid cultures typically lack resident immune cells, such as macrophages, dendritic cells, and T cells. This absence severely limits their ability to model critical processes such as immune-mediated toxicity, responses to infection, inflammation, and tissue repair, which are central to diseases such as cancer, autoimmune disorders, and infectious diseases [
93].
Organoids often fail to develop a functional circulatory system. The resulting deficits in nutrient delivery, oxygenation, and waste removal restrict their growth potential, long-term viability, and functional maturity. This is particularly problematic for modeling chronic drug exposure or diseases where vascular interactions are key, and prevents accurate pharmacokinetic or pharmacodynamic modeling of drug distribution [
94,
95].
While organoids replicate primary parenchymal cells, they frequently lack the full complement of supportive cell types found in the native niche. This simplified cellular composition disrupts authentic cell-cell signaling, tissue organization, and functional maturity, hindering the modeling of complex processes such as fibrosis, regeneration, and neural modulation [
96].
These inherent gaps—the lack of immune response, vascularization, and full cellular diversity—collectively affect the accuracy of drug testing. They can lead to incomplete replication of in vivo pharmacokinetics and an underestimation of systemic and immune-mediated toxicity, ultimately reducing the predictive value of conventional organoid models.
5.3. Engineering strategies to overcome functional limitations
These limitations are being actively addressed through innovative engineering and biological strategies that aim to create next-generation organoid systems with enhanced physiological relevance. These strategies directly target the gaps in immunity, vascularization, and complexity. A primary criticism is that many organoid models remain sophisticated avatars rather than true functional proxies for human organs.
The common absence of immune components creates a biologically unrealistic sanctuary, failing to capture critical drug-induced immune responses such as idiosyncratic hepatotoxicity or cytokine release syndrome, which are major causes of drug failure [
93]. To address the lack of an immune system, researchers are developing immune‒organoid cocultures. This involves incorporating derived macrophages, microglia (in neural organoids), or other immune cells into the organoid culture system [
87]. Furthermore, genetically modifying organoids to express human leukocyte antigens (HLAs) or using CRISPR/Cas9 to introduce immune checkpoints can better model immune-oncology interactions and immune-mediated toxicity. These approaches allow the study of neuroinflammation, immune-related adverse events (irAEs), and personalized cancer immunotherapy.
The frequent lack of a perfusable vascular system is not merely a technical hurdle; it fundamentally constrains nutrient diffusion, limits size, and disrupts the gradient-dependent signaling crucial for zonation in organs such as the liver, thus skewing metabolic studies [
94,
95]. The vascularization challenge is met through multiple engineering approaches. Microfluidic OOC platforms provide dynamic perfusion, simulate blood flow, and improve nutrient delivery to sustain larger and more mature organoids [
90]. 3D bioprinting allows for the precise patterning of endothelial cells and bioinks to create predesigned, perfusable vascular networks within organoids [
84,
89]. Additionally, coculturing organoids with endothelial cells promotes the self-formation of capillary-like structures
in vitro, which can often anastomose with host vasculature upon implantation [
88].
To achieve more complex microenvironments, advanced coculture systems are being employed to incorporate fibroblasts, stromal cells, and neural cells alongside primary organoid cells. These better mimic the
in vivo niche and promote maturation. Bioreactor cultures provide mechanical and biochemical cues that are essential for driving the functional maturity of tissues such as muscle, heart, and cartilage [
80,
83]. These systems promote better tissue organization and more adult-like phenotypes. Acknowledging these gaps not only tempers the interpretation of organoid data but also provides a critical roadmap for prioritizing engineering efforts, such as vascularization and immune integration, to construct more holistic systems.
5.4. Barriers to widespread use of organoid technology
While organoid technology faces significant cost, standardization, and scalability barriers, advancements in automation, microfluidics, and bioreactor systems are addressing these challenges [
97,
98]. Simplified protocols and standardized tools will increase affordability and accessibility, paving the way for broader adoption in drug testing, disease modeling, and personalized medicine.
Organoid cultures rely on expensive reagents, such as ECM components and growth factors, which are particularly costly for large-scale experiments or high-throughput screening [
99]. Specialized infrastructure bioreactors, incubators, imaging equipment, and microfluidic devices add to the expense. These systems require skilled personnel, extensive training, and regular maintenance, further increasing operational costs. This limits accessibility to well-funded labs, restricting adoption in smaller or resource-limited settings.
The lack of standardized, affordable kits for specific organs or tissues forces researchers to develop custom protocols, increasing complexity and labor costs. Custom approaches complicate data interpretation and make cross-lab comparisons challenging. The absence of off-the-shelf solutions hinders widespread adoption in drug testing and clinical applications.
The large-scale production of organoids is imperative for reducing expenses and facilitating more widespread clinical utilization [
19]. This approach has been shown to reduce labor costs, improve consistency, and enhance throughput for large-scale drug screening. Furthermore, the capacity for simultaneous testing of multiple organoids on a single chip is enhanced, thus improving efficiency and reducing reagent waste. It facilitates the controlled, large-scale production of consistent organoid batches, thus reducing per-unit costs.
Efforts to simplify protocols such as using less expensive reagents or synthetic scaffolds are underway to reduce costs. Collaborative initiatives, such as those led by the Human Cell Atlas, aim to standardize organoid generation and testing methods globally [
100]. Standardized kits and protocols lower barriers to entry, improve reproducibility, and make organoid technology more accessible (
Table 4 [
17], [
77], [
80], [
83], [
85], [
86], [
87],[
93], [
94], [
95], [
96], [
97], [
98], [
99],[
101], [
102], [
103], [
104], [
105], [
106], [
107], [
108], [
109], [
110], [
111]).
6. Ethical and regulatory considerations
6.1. Ethical considerations in organoid research
The rapid advancement of organoid technology necessitates a robust and nuanced ethical framework. The ethical landscape can be broadly divided into two primary categories: ① the ethics of source material and donor consent, which applies to all organoid types, and ② the unique ethical implications arising from the increasing cellular and functional complexity of certain organoids, particularly those modeling neural tissues.
The generation of organoids, especially PDOs, raises fundamental questions regarding donor rights and tissue autonomy [
102]. The procurement of starting material, typically via biopsies or surgical waste, requires a robust informed consent process that is often inadequately addressed in broad institutional consent forms. Donors may not fully anticipate or consent to the wide range of potential future uses of their cells, including prolonged
in vitro culture, genetic manipulation, commercialization, and sharing with third-party researchers or industry partners. Furthermore, PDOs are frequently used for genetic and molecular profiling, resulting in significant privacy concerns [
103]. The sensitive genetic information derived from these models could be linked back to the donor, with implications for insurance, employment, and familial relationships.
To address these issues, leading guidelines, such as those from the International Society for Stem Cell Research (ISSCR), recommend implementing tiered consent processes [
104]. This allows donors to specify their preferences for specific research applications, future commercial use, and data sharing. Additionally, strict protocols for anonymization and data protection, in compliance with regulations such as the General Data Protection Regulation (GDPR) in Europe and the Health Insurance Portability and Accountability Act (HIPAA) in the United States, are considered mandatory best practices to safeguard donor privacy and autonomy.
A second, more speculative set of ethical concerns arises as organoids become more advanced, particularly in the context of neural models involving neural organoids and consciousness. The primary ethical debate surrounds the possibility that highly complex neural organoids could develop some form of consciousness or sentience capable of experiencing pain or distress. While the current scientific consensus, which is based on structural and functional evidence, holds that today’s brain organoids are not conscious, their potential to exhibit neural activities associated with awareness demands proactive ethical stewardship [
105,
106]. The ISSCR guidelines recommend that research involving neural organoids that incorporate sensory-like inputs or are transplanted into animal models receive specialized oversight from expanded ethics committees that include neuroscientists, ethicists, and legal experts [
104,
107]. The goal is not to halt progress but to establish a framework for proportional review, where the level of oversight is matched to the model’s complexity and potential capabilities.
Moral status and research protections lead to the deeper philosophical question of the moral status of a conscious organoid. Should such an entity be afforded protections similar to those granted to research animals? Current regulations do not address this possibility. The National Institutes of Health (NIH) maintains a cautious stance on research that could confer human-like attributes to animal models, a principle that can be extended to highly advanced human organoids [
104]. The ethical imperative is to define empirical benchmarks for concerning neural capabilities to guide future oversight, moving the discourse from speculation to evidence-based policy.
6.2. Regulatory perspectives
As organoid technologies enter clinical settings, clear and standardized regulatory frameworks are essential. Regulatory bodies such as the FDA, the European Medicines Agency (EMA), and International Organization for Standardization (ISO) are starting to develop validation criteria for organoid models [
107], [
108], [
109]. These criteria cover biological relevance, reproducibility, and ethical aspects. ISO guidelines, for example, aim to set international standards for organoid reproducibility, quality, and functionality in organ-specific modeling, drug screening, and personalized medicine [
110]. Clear regulatory pathways for organoid-based therapies need to be established, including clinical trial design, safety testing, and approval processes. Close collaboration among regulatory agencies, academia, and industry is crucial to create a unified approach for organoid technology approval. Standardized guidelines and regulatory clarity will facilitate the translation of organoid research into clinical applications and attract industry investment.
The principal ethical concern for neural organoid research is the potential for advanced models, particularly those derived from human iPSCs, to exhibit rudimentary forms of consciousness or sentience. As these organoids develop increasingly complex neural networks and exhibit electrical activity resembling that of the developing human brain, including synchronized bursting and synaptic plasticity, speculative questions arise regarding their capacity for pain perception or conscious experience [
105,
112]. While the current scientific consensus holds that today’s organoids are not conscious, the rapid pace of technological advancement necessitates proactive ethical stewardship [
113].
In response, leading international organizations have begun to establish guidelines for this research. The ISSCR has updated its guidelines to specifically address human embryo and organoid research, recommending heightened oversight for studies that could confer aspects of consciousness. These guidelines suggest limitations on gestational age
in vitro, restrict the transplantation of human neural organoids into animal models, and call for the establishment of specialized oversight committees to review such sensitive research protocols [
104,
107]. Similarly, the NIH advises a cautious approach, particularly concerning neural chimera research, emphasizing the principle of proportionality where the level of oversight is matched to the perceived ethical risk [
114]. Therefore, moving forward, rigorous ethical oversight by institutional review boards (IRBs) and dedicated bioethics committees is not just recommended but essential. Research protocols involving neural organoids must be subject to regular review, ensuring transparent conduct and adherence to the evolving international framework that aims to balance scientific innovation with profound ethical responsibility [
17,
111].
6.3. Need for international harmonization and validation standards
The translational pathway for organoid-based applications is fraught with regulatory uncertainty, as global agencies grapple with how to classify and evaluate these complexes, living products. While regulatory bodies acknowledge the potential of organoids to improve predictive drug development, the absence of formalized, specific guidelines creates a significant barrier to their adoption in regulatory decision-making. A comparative analysis reveals both converging principles and distinct approaches among leading agencies.
The FDA has demonstrated a proactive, pilot-based approach to innovation. This is exemplified by the Innovative Science and Technology Approaches for New Drugs (ISTAND) Pilot Program, which explicitly considers novel tools such as organoids for potential qualification as drug development tools (DDTs). The FDA Modernization Act 2.0 of 2022, which removed the mandatory requirement for animal testing for new drugs, further created a regulatory environment conducive to accepting human-relevant models such as organoids [
18]. The FDA’s evolving stance emphasizes a context-of-use framework, where the acceptability of organoid data depends on the specific question being asked (e.g., mechanistic toxicity and specific organ toxicity) and the strength of the validation package provided.
The EMA and the European Union (EU)’s governing bodies are pursuing a more structured, guideline-driven pathway. The EMA has established procedures for the qualification of novel methodologies, which could provide a route for organoid model qualification. A significant development is the EU Delegated Regulation 2023, which, in alignment with the EU’s Chemicals Strategy for Sustainability, aims to phase out animal testing and promotes the use of new approach methodologies (NAMs), a category that includes organoids and organ-on-chip models. This creates a top-down regulatory push for alternatives, although specific technical guidelines for organoids are still under development.
The ISO plays a complementary role by focusing on technical standardization and quality control. While not a regulatory body, the ISO develops international consensus standards that provide critical benchmarks for reproducibility, characterization, and functionality [
110]. These standards are being increasingly referenced by regulatory agencies such as the FDA and the EMA as they seek to establish baseline quality requirements for submitted data. The ISO’s work is crucial for creating the common language and technical specifications needed for global regulatory harmonization.
Despite these initiatives, a significant gap remains. Pharmaceutical companies lack clear guidance on the validation standards required for organoid-based data to support the Investigational New Drug (IND) or the New Drug Application (NDA) submissions. The critical need is for collaboration between academic researchers, industry stakeholders, and regulatory bodies to generate the robust, multilaboratory validation data required for formal guideline development. Initiatives led by the ISSCR are working toward best practices, but formal agency-specific guidelines defining suitable organoid models, validation standards, and acceptable data end points are the essential next step for the field.
In parallel with the FDA and the EMA, China’s Center for Drug Evaluation (CDE) under the National Medical Products Administration (NMPA) has established a comprehensive regulatory framework for the evaluation of innovative drugs and advanced therapies. The CDE is responsible for the scientific and technical review of the IND applications, focusing on pharmacology, toxicology, clinical pharmacokinetics, and efficacy assessments. In recent years, the CDE has released guidelines addressing cell and gene therapies, regenerative medicine products, and complex
in vitro models, reflecting its strategic emphasis on translational relevance and risk-benefit evaluation. The inclusion of CDE perspectives is particularly pertinent, as China is emerging as a major contributor to organoid and tissue engineering research, and its regulatory standards increasingly influence global harmonization efforts (
Table 5).
7. Future directions
7.1. Technological innovations
Artificial intelligence (AI) and machine learning (ML) are being increasingly incorporated into organoid research [
92,
115]. They can analyze large datasets from organoid-based drug screenings. ML models can study data from organoid cultures exposed to different drugs and detect patterns in gene expression, cellular behavior, and metabolic changes related to known toxicity profiles [
54]. AI can also predict drug candidate efficacy by analyzing organoid responses at the cellular and molecular levels across different organoid models [
116]. This helps in personalized treatment approaches, using PDOs to screen drug responses. However, integrating diverse data types into AI models is challenging, and standardized datasets and multiomics data integration are needed for better accuracy. Regulatory agencies must also set guidelines for using AI in predictive modeling for drug safety and efficacy testing.
Vascularization and innervation are crucial for functional organs but are often lacking in organoid models [
94,
96,
117]. Coculturing organoids with endothelial and neural cells is a growing strategy. Coculturing with endothelial cells can generate more realistic, vascularized organoid models, improving the nutrient and oxygen supply for large or long-term treated organoids. These can also assess drug effects on the vascular system. Adding neural cells to organoid cultures allows the study of organ-specific neural functions. In neural organoids, it helps model the blood-brain barrier and assess neurotoxicity [
118]. Combining both cell types with organoids enables more accurate disease modeling for conditions such as stroke and neurodegenerative disorders [
36]. However, this requires precise control over culture conditions to optimize cell‒cell interactions and establish functional networks, which is a technical challenge for scaling up.
CRISPR/Cas9 gene editing is a powerful tool for generating disease-specific mutations in organoids [
119]. It allows researchers to generate organoid models that closely mimic genetic mutations in diseases such as cystic fibrosis or Parkinson’s disease [
120]. These models are used for drug screening to identify therapies targeting genetic causes. In cancer research, CRISPR-edited organoids help in the identification of drugs for tumor mutation [
121]. It is also used in regenerative medicine to edit stem cells in organoids [
122]. However, off-target effects can occur and ethical questions arise, especially regarding germline cell editing. Ethical guidelines are needed, and regulatory agencies must set rules for the safe and ethical use of CRISPR in organoid-based research, particularly in clinical applications.
7.2. Integration with regenerative therapies
The integration of organoids with biomaterials and growth factors represents a transformative approach to regenerative medicine, enabling the creation of functional, durable, and patient-specific tissue implants [
123]. While challenges remain, advancements in vascularization, immune modulation, and controlled growth factor delivery hold promise for addressing unmet clinical needs and improving outcomes in tissue repair and regeneration.
The creation of organoid-derived implants, such as bioengineered liver patches [
124], is an exciting development in organoid-based therapies, especially for treating acute liver failure (ALF). ALF is life-threatening, and because of donor organ shortages and transplantation complications, alternative therapies are needed. Liver patches, made from liver organoids derived from human stem cells or primary liver cells, are designed to mimic liver tissue functions such as detoxification, protein synthesis, and metabolic regulation. They can be implanted to support liver function temporarily while waiting for a transplant or as a long-term solution. The patches aim to integrate with native tissue and secrete growth factors for liver regeneration. However, challenges include ensuring sufficient vascularization and minimizing immune rejection.
Combining organoids with biomaterial scaffolds and growth factors is a powerful approach for tissue regeneration. Biomaterial scaffolds such as hydrogels or decellularized ECM provide structural support, maintain organoid shape, and aid in vascularization and cellular infiltration. Cardiac organoids on heart ECM-mimicking scaffolds promote cell function [
125], and liver organoids on suitable scaffolds support long-term metabolic activity [
126,
127]. Growth factors such as vascular endothelial growth factor (VEGF), epidermal growth factor (EGF), and fibroblast growth factor (FGF) enhance organoid growth, differentiation, and maturation within the scaffold. They stimulate tissue development, improve cellular communication, and are crucial for vascularization [
95,
128]. In liver organoids, VEGF can increase blood vessel formation for better host-tissue integration. This combination creates a dynamic environment for tissue repair, simulating the
in vivo microenvironment to promote better tissue organization [
129]. It allows for patient-specific implants, accelerating healing and reducing the reliance on organ transplantation. However, challenges exist, as biomaterial scaffolds must be biocompatible, offer proper mechanical support and degradability, and be porous for cell and vessel growth. Additionally, efficiently and controllably delivering growth factors over a sustained period is difficult, as improper delivery can affect tissue function.
7.3. Bridging the gap between organoid research and clinical applications
The scalability of organoid production is imperative for reducing costs and facilitating broader clinical adoption. Automated bioreactor systems have been demonstrated to decrease labor costs, increase consistency, and increase throughput for large-scale drug screening applications [
130] (
Table 6,
Table 7,
Table 8,
Table 9,
Table 10,
Table 11,
Table 12,
Table 13). Furthermore, microfluidic platforms enable the simultaneous culture and analysis of multiple organoids on a single chip, thus improving experimental efficiency and reducing reagent consumption. These technologies also support the controlled, large-scale production of uniform organoid batches, effectively reducing per-unit costs.
Case studies of organoid-guided therapies offer insights into their practical clinical use. To treat liver failure, organoid-derived hepatocytes (eLOs) have been used. When generated from liver organoids, eLOs can be transplanted into patients with liver failure [
131]. Clinical trials with patient-specific eLOs have shown potential for regenerating liver tissue, supporting detoxification, and restoring function, either as a bridge to transplantation or as an alternative in some cases [
132]. However, challenges remain, such as the risk of immune rejection (even with patient-derived cells) and ensuring long-term viability and function of transplanted cells [
133].
7.4. Revolutionizing TERM therapies with organoids
Organoids are revolutionizing TERM therapies by enabling personalized drug testing and regenerative treatments [
22,
134]. Their ability to replicate human biology and integrate with advanced technologies ensures safer, more effective therapies, ultimately advancing precision medicine and improving patient outcomes.
Personalized medicine aims to customize treatment according to a patient’s genetic makeup, disease profile, and drug response. PDOs offer an innovative way to model a patient’s disease and test drug safety and efficacy in a human-relevant system. In drug discovery and toxicity testing, PDOs are generated from patient tissues such as tumor biopsies [
135], liver samples [
136], or skin cells [
28]. For cancer treatment, tumor-derived PDOs can be exposed to chemotherapy drugs to predict patient response and drug toxicity [
15]. In cardiovascular disease, patient-specific heart organoids help assess drug effects on heart tissue [
137]. In oncology, PDOs screen different chemotherapies and targeted therapies, identify drug resistance mechanisms, and develop personalized cancer treatment strategies [
102]. The incorporation of genomic analysis into PDO models in pharmacogenomics helps link genetic variations to drug responses and the identification of biomarkers. PDOs also enable fine-tuning of drug dosing, reducing adverse effects and leading to precision prescription [
138].
Organoids are crucial for personalized regenerative medicine. In heart repair after myocardial infarction, patient-derived cardiac organoids can be used to test stem cell-based therapies [
24,
139]. In liver diseases such as cirrhosis or ALF, liver organoids can be used to model regeneration and test stem cell-based therapies [
140,
141]. With respect to neurodegenerative diseases such as Alzheimer’s disease or Parkinson’s disease, PDOs can be used to study disease progression and test regenerative therapies [
22,
142]. Patient-specific organoids can generate tissues for autologous transplants [
143]. Since they are from the patient’s own cells, there is a reduced risk of immune rejection compared with traditional organ transplants, increasing the success rate of therapies by tailoring to the patient’s unique characteristics.
8. Concluding remarks
Organoids represent a transformative modality in the field of drug evaluation, providing human-specific insights that deepen our understanding of tissue function, disease modeling, and therapeutic responses. By recapitulating the architectural and functional complexity of human organs in vitro, organoid models yield more accurate and physiologically relevant data than traditional 2D cultures or animal models do. Their ability to model diverse tissues ranging from liver and neural to cardiac and vascular systems makes them indispensable tools in the advancing fields of regenerative medicine and tissue engineering. These models help bridge the translational gap between preclinical studies and clinical application, promoting the development of precisely targeted therapies and improved safety profiles for patients. Despite their considerable potential, several challenges must be addressed to enable the widespread adoption of organoid technologies. Key issues include the standardization of protocols to ensure reproducibility across different laboratories and platforms and the resolution of ethical concerns related to PDO models, including informed consent, privacy, and the use of genetic data. The future of organoid-based drug testing is highly promising, particularly for advancing tissue engineering and regenerative therapies. The integration of organoid technology into routine clinical workflows from drug screening to patient-specific therapeutic interventions paves the way for a new era of medicine that is more personalized, effective, and safe.
PDF
(2248KB)
