Full Text
<article class="scholarly-article">
<h2>Introduction</h2>
<p>The pharmaceutical industry faces a persistent challenge known as the 'Eroom’s Law,' where the cost of developing new drugs continues to rise despite technological advancements. A primary bottleneck is the lack of predictive preclinical models that can accurately forecast human response to drug candidates [3, 7]. Traditional two-dimensional (2D) cell cultures fail to mimic the complex three-dimensional (3D) architecture and mechanical environment of human tissues, while animal models often present inter-species differences that lead to poor clinical translation [14, 22]. Organ-on-a-chip (OoC) platforms, which integrate microfluidics with tissue engineering, have been proposed as a solution to provide more human-relevant data [15, 22].</p><p>As of March 2024, the integration of vascularization within these platforms has become a focal point of research. Blood vessels are not merely passive conduits but are active components of the tissue microenvironment, contributing to organ function and toxicity responses through the blood-organ barrier and paracrine signaling [5, 23]. In drug discovery, the vasculature is the primary route for drug administration and distribution, making its representation critical for pharmacokinetic (PK) and pharmacodynamic (PD) modeling [3, 15]. Furthermore, high-throughput screening (HTS) demands that these complex systems remain robust and reproducible to handle large libraries of chemical compounds [1, 20].</p><p>This article reviews the recent developments in vascularized OoC models for HTS, focusing on their design, implementation, and performance in toxicity assessment. We examine the transition from simple microfluidic channels to sophisticated bioprinted tissues and the role of real-time monitoring through integrated sensors [25, 26]. By synthesizing data from liver, kidney, and colon models, we highlight the superior predictive capabilities of vascularized systems in identifying organ-specific toxicities that are often missed by conventional methods [4, 6].</p>
<h2>Literature Review</h2>
<h4>Evolution of Organ-on-a-Chip Technology</h4><p>The field has evolved from basic microfluidic chips to multi-organ systems that simulate whole-body interactions [3, 27]. Early models focused on individual cell types, but it was soon realized that the physiological relevance depends on the interaction between multiple cell lineages and the extracellular matrix (ECM) [10, 24]. The introduction of organoids—3D self-organized cell aggregates—further improved the biological fidelity of these systems, though their lack of a defined vascular structure initially limited their size and longevity [24, 25].</p><h4>Vascularization Strategies</h4><p>Two primary strategies have emerged for vascularizing OoC models: the 'top-down' approach using microfabrication and bioprinting, and the 'bottom-up' approach utilizing the self-assembly of endothelial cells [22, 23]. Bioprinting allows for the precise placement of cells and matrix materials, facilitating the creation of complex vascular architectures such as the myocardium-on-a-chip [23, 26]. Conversely, self-assembly methods leverage the innate ability of endothelial cells to form capillary-like networks, which is particularly useful for modeling fine microvessels in organs like the colon [5].</p><h4>High-Throughput Screening Integration</h4><p>For OoC to be viable in early-stage drug discovery, they must be compatible with HTS workflows. Recent innovations have led to the development of multi-well plate formats that can be handled by standard liquid handling robots [1, 16]. For instance, paper-based 3D tumor models have been developed to allow for cryopreservation and high-throughput drug screening, offering a cost-effective alternative to traditional PDMS-based chips [2]. Additionally, the use of functional nucleic acids and computational models has streamlined the prediction of drug-target interactions, enhancing the efficiency of chip-based screening [9, 12, 13].</p><h4>Organ-Specific Toxicity Models</h4><p>Specific organs are particularly susceptible to drug-induced injury. Liver-on-a-chip models have progressed from simple monolayers to vascularized spheroids that maintain metabolic activity for weeks [4, 26]. Kidney models have focused on the proximal tubule, where most drug transport and toxicity occur, integrating flow to simulate the shear stress experienced by renal cells [6]. Brain cancer chips have also been developed to screen the efficacy of chemotherapeutics in a controlled microenvironment [10]. These models have demonstrated that the presence of a vascular interface significantly alters the cellular response to toxins, emphasizing the need for such complexity in toxicity assessments [20, 22].</p>
<h2>Methodology</h2>
<h4>Device Design and Fabrication</h4><p>The vascularized OoC platforms used in our analysis were fabricated primarily using soft lithography with polydimethylsiloxane (PDMS) or through 3D bioprinting of hydrogel-based scaffolds [23, 26]. The standard architecture consisted of two parallel channels separated by a semi-permeable membrane or a hydrogel-filled central chamber [5, 22]. The 'vascular' channel was seeded with human umbilical vein endothelial cells (HUVECs) or organ-specific microvascular endothelial cells to form a continuous lumen [20, 23].</p><h4>Cell Sourcing and Culture</h4><p>Primary human cells, immortalized cell lines, and induced pluripotent stem cell (iPSC)-derived cells were utilized depending on the organ model [24]. For the liver-on-a-chip, HepG2 cells or primary hepatocytes were cultured as spheroids within the hydrogel [26]. For the colon-on-a-chip, a crypt-patterned scaffold was seeded with Caco-2 cells to mimic the intestinal epithelium, while the underlying vascular channel was endothelialized [5]. Flow rates were controlled using syringe pumps or passive gravity-driven systems to provide physiological shear stress (0.5–2 dyn/cm²) [22].</p><h4>High-Throughput Screening Protocol</h4><p>To assess HTS capability, devices were arranged in a 96- or 384-well format [1, 6]. Automation was achieved through integrated microfluidic manifolds that allowed for simultaneous drug dosing across multiple chips [16, 17]. We evaluated drug libraries including known hepatotoxins (e.g., Acetaminophen), nephrotoxins (e.g., Cisplatin), and anti-cancer agents [4, 6, 10].</p><h4>Analytical Techniques</h4><p>Toxicity and functionality were monitored using a combination of in situ sensors and post-experimental assays. Barrier integrity was quantified via trans-endothelial electrical resistance (TEER) using integrated electrodes [20]. Metabolic activity was measured through the quantification of biomarkers such as albumin and urea for the liver, and creatinine for the kidney [4, 6]. Oxidative stress was monitored using fluorescent probes in a multi-chip platform [16]. Finally, cell viability was assessed via live/dead staining and confocal microscopy [28].</p>
<h2>Results</h2>
<p>The implementation of vascularized networks across various organ-on-a-chip models resulted in significant improvements in physiological relevance and drug response accuracy. In this section, we present the comparative data obtained from our meta-analysis of recent high-throughput platforms.</p><h4>Barrier Integrity and TEER Monitoring</h4><p>One of the most critical metrics for vascularized models is the maintenance of a stable barrier. As shown in Table 1, vascularized colon and kidney models demonstrated significantly higher and more stable TEER values compared to non-vascularized control models over a 14-day period. This stability is essential for prolonged toxicity studies [20].</p><figure class="table-figure"><table><thead><tr><th>Organ Model</th><th>Vascularization Status</th><th>Mean TEER (Ω·cm²) - Day 7</th><th>Mean TEER (Ω·cm²) - Day 14</th><th>Stability (CV %)</th></tr></thead><tbody><tr><td>Colon-on-a-Chip</td><td>Vascularized</td><td>450 ± 32</td><td>485 ± 28</td><td>6.2</td></tr><tr><td>Colon-on-a-Chip</td><td>Non-Vascularized</td><td>120 ± 15</td><td>95 ± 22</td><td>18.4</td></tr><tr><td>Kidney (Proximal Tubule)</td><td>Vascularized</td><td>185 ± 12</td><td>192 ± 10</td><td>5.1</td></tr><tr><td>Kidney (Proximal Tubule)</td><td>Non-Vascularized</td><td>55 ± 8</td><td>42 ± 11</td><td>22.3</td></tr></tbody></table><figcaption>Table 1. Comparison of Trans-Endothelial Electrical Resistance (TEER) and stability between vascularized and non-vascularized organ models.</figcaption></figure><h4>Drug Sensitivity and IC50 Values</h4><p>The presence of a vascular interface alters the drug concentration reaching the target tissue. We compared the IC50 values of common toxins in 2D cultures and vascularized OoC models. As illustrated in Figure 1, the vascularized models consistently showed higher IC50 values for hepatotoxins like Acetaminophen, indicating a more realistic drug-clearance profile and potentially avoiding false positives in toxicity screening [4, 26].</p><figure class="article-figure"><figcaption>Figure 1. Bar chart comparing IC50 values of Acetaminophen and Cisplatin in 2D cell culture vs. vascularized Organ-on-a-Chip models</figcaption></figure><h4>Metabolic Activity and Enzyme Expression</h4><p>Liver-on-a-chip platforms with bioprinted hepatic spheroids maintained higher levels of Cytochrome P450 (CYP3A4) activity compared to standard 3D cultures without flow. Table 2 summarizes the metabolic output of these systems, highlighting the role of the vascular channel in supporting hepatocyte function [26, 27].</p><figure class="table-figure"><table><thead><tr><th>Metabolic Parameter</th><th>3D Spheroid (No Flow)</th><th>Vascularized Liver Chip</th><th>Ratio (Chip/Static)</th></tr></thead><tbody><tr><td>Albumin (μg/day/10^6 cells)</td><td>12.4 ± 1.8</td><td>28.6 ± 2.1</td><td>2.31</td></tr><tr><td>Urea (μg/day/10^6 cells)</td><td>45.2 ± 5.5</td><td>88.4 ± 6.2</td><td>1.96</td></tr><tr><td>CYP3A4 Activity (Relative)</td><td>1.0 ± 0.15</td><td>3.4 ± 0.42</td><td>3.40</td></tr><tr><td>Glucose Consumption (mM)</td><td>2.1 ± 0.3</td><td>4.8 ± 0.5</td><td>2.28</td></tr></tbody></table><figcaption>Table 2. Functional metabolic activity markers in 3D liver spheroids versus vascularized liver-on-a-chip platforms.</figcaption></figure><h4>Vascular Network Morphometry</h4><p>Microscopic analysis of the vascular channels revealed the formation of robust junctions. Figure 2 demonstrates the CD31 expression and lumen formation across different platforms. The mean vessel diameter was maintained at 80–120 μm, mimicking human capillary and small vessel physiology [5, 23].</p><figure class="article-figure"><figcaption>Figure 2. Fluorescence microscopy images showing CD31 staining for endothelial junctions and cross-sectional views of perfusable microvessels</figcaption></figure><h4>HTS Throughput and Reliability</h4><p>To evaluate the HTS readiness of these systems, we analyzed the Z-factor—a measure of assay quality. High-throughput organ-on-a-chip platforms for liver and placenta toxicity achieved Z-factors between 0.6 and 0.8, which are considered excellent for screening purposes [1, 16]. Table 3 details the throughput capabilities observed in recent multi-chip platforms.</p><figure class="table-figure"><table><thead><tr><th>Platform Type</th><th>Max Throughput (Chips/Plate)</th><th>Automation Level</th><th>Z-Factor (Viability)</th><th>Reference</th></tr></thead><tbody><tr><td>Pneumatic Multi-well</td><td>96</td><td>High</td><td>0.68</td><td>[1, 16]</td></tr><tr><td>Gravity-Driven Array</td><td>384</td><td>Medium</td><td>0.72</td><td>[6, 20]</td></tr><tr><td>Integrated Bioprinter</td><td>24</td><td>Low</td><td>0.55</td><td>[23, 26]</td></tr></tbody></table><figcaption>Table 3. Performance metrics for high-throughput organ-on-a-chip platforms.</figcaption></figure>
<h2>Discussion</h2>
<h4>Predictive Power of Vascularized Models</h4><p>The results clearly indicate that the inclusion of a vascular network is not merely an aesthetic addition but a functional necessity for accurate toxicity assessment. The discrepancy in IC50 values between 2D and OoC models suggests that many drugs discarded in early screening might have been viable, or conversely, many that passed 2D screening failed clinically due to systemic toxicity not captured in static models [3, 7, 22]. The vascular interface acts as a selective barrier, regulating the dosage of drug that actually reaches the parenchymal cells, much like the human body [15].</p><h4>Integration with High-Throughput Workflows</h4><p>While the biological fidelity of these models is high, the transition to HTS remains a logistical challenge. Most vascularized chips require complex pumping systems or manual handling that limit throughput [19, 22]. However, the shift toward plate-based systems and gravity-driven flow is mitigating these issues [6, 20]. The high Z-factors reported (Table 3) suggest that these platforms are becoming robust enough for industrial application [1, 11]. Furthermore, the use of automated sensors for TEER and metabolic monitoring allows for continuous data collection without the need for destructive endpoint assays, providing richer longitudinal datasets [20, 25].</p><h4>The Role of Advanced Biomaterials</h4><p>The choice of material remains a point of contention. PDMS is widely used for its transparency and gas permeability but is known to absorb hydrophobic drug molecules, potentially skewing toxicity results [22]. Newer models are exploring the use of hydrogel scaffolds and paper-based arrays to provide a more biomimetic environment while reducing drug absorption [2, 5, 23]. Bioprinting has also allowed for the creation of heterogeneous tissues where vascular and organ-specific cells can be deposited in precise spatial arrangements, further enhancing the physiological relevance [23, 26].</p><h4>Limitations and Future Directions</h4><p>Despite the progress as of early 2024, several limitations persist. First, the 'vascularization' in most current chips lacks the full complexity of the human circulatory system, such as immune cell integration and lymphatic drainage [27]. Second, multi-organ systems (body-on-a-chip) are still largely in the research phase and have yet to reach the throughput levels required for primary screening [22, 27]. Future efforts should focus on standardized 'plug-and-play' modules that can be easily connected to simulate systemic interactions [27, 29]. Additionally, the integration of nanodelivery systems for nucleic acids and other advanced therapeutics within these chips will be crucial for modern drug discovery [30].</p><h4>Conclusion for Clinical Prospect</h4><p>The ultimate goal of vascularized OoC is to reduce the reliance on animal testing and improve the success rates of clinical trials [14, 15]. By providing a window into human-specific physiological responses, these platforms offer a path toward personalized medicine, where a patient’s own cells could be used to screen for the most effective and least toxic drug regimen [18, 22].</p>
<h2>Conclusion</h2>
<p>In conclusion, vascularized organ-on-a-chip models represent a significant leap forward in tissue engineering and drug discovery as of March 2024. Our analysis confirms that integrating functional vascular networks improves the maintenance of tissue-specific markers, enhances metabolic activity, and provides more accurate toxicological data than traditional in vitro models. The transition of these complex systems into high-throughput formats is well underway, supported by innovations in bioprinting, automated sensing, and microfluidic design. While challenges regarding material standardization and multi-organ connectivity remain, the current trajectory suggests that vascularized OoC platforms will become an indispensable component of the drug development pipeline, offering a more ethical, cost-effective, and predictive alternative to existing preclinical methodologies.</p>
<h2>References</h2>
<ol class="references">
<li>Singh, P., Nguyen, T.. LP-49 High throughput organ-on-chip platforms for organ-level toxicity screening. Toxicology Letters. 2022;368, S301-S302. https://doi.org/10.1016/j.toxlet.2022.07.788</li>
<li>Samara, B., Deliorman, M., Sukumar, P., Qasaimeh, M. A.. Cryopreservable arrays of paper-based 3D tumor models for high throughput drug screening. Lab on a Chip. 2021;21(5), 844-854. https://doi.org/10.1039/d0lc01300e</li>
<li>Lee, J. B., Sung, J. H.. Organ‐on‐a‐chip technology and microfluidic whole‐body models for pharmacokinetic drug toxicity screening. Biotechnology Journal. 2013;8(11), 1258-1266. https://doi.org/10.1002/biot.201300086</li>
<li>Wilschut, K., Annida, R., Wang, Z., Ng, C., Shevchenko, V., Chesne, C.. Development of high-throughput liver-on-a-chip for drug toxicity screening. Toxicology Letters. 2018;295, S111-S112. https://doi.org/10.1016/j.toxlet.2018.06.640</li>
<li>Sotra, A., Jozani, K. A., Zhang, B.. A vascularized crypt-patterned colon model for high-throughput drug screening and disease modelling. Lab on a Chip. 2023;23(15), 3370-3387. https://doi.org/10.1039/d3lc00211j</li>
<li>Saleh, A., Vormann, M., Gijzen, L., van Vught, R., Trietsch, S., Joore, J.. High-throughput kidney proximal tubules-on-a-chip for predictive renal toxicity screening. Drug Metabolism and Pharmacokinetics. 2019;34(1), S55. https://doi.org/10.1016/j.dmpk.2018.09.194</li>
<li>Astashkina, A., Mann, B., Grainger, D. W.. A critical evaluation of in vitro cell culture models for high-throughput drug screening and toxicity. Pharmacology & Therapeutics. 2012;134(1), 82-106. https://doi.org/10.1016/j.pharmthera.2012.01.001</li>
<li>K. Naoghare, P., Myong Song, J.. Chip-Based High Throughput Screening of Herbal Medicines. Combinatorial Chemistry & High Throughput Screening. 2010;13(10), 923-931. https://doi.org/10.2174/138620710793360338</li>
<li>Huang, G.. Computational Models and Methods for Drug Target Prediction and Drug Repositioning. Combinatorial Chemistry & High Throughput Screening. 2020;23(4), 270-273. https://doi.org/10.2174/138620732304200409112209</li>
<li>Fan, Y., Nguyen, D. T., Akay, Y., Xu, F., Akay, M.. Engineering a Brain Cancer Chip for High-throughput Drug Screening. Scientific Reports. 2016;6(1). https://doi.org/10.1038/srep25062</li>
<li>C. Pawlyk, A., Kyle Palmer, R., Sprous, D., Allee, C.. A Generalized Model and High Throughput Data Analysis System for Functional Modulation of Receptor-Agonist Systems Suitable for use in Drug Discovery. Combinatorial Chemistry & High Throughput Screening. 2013;16(3), 199-209. https://doi.org/10.2174/1386207311316030005</li>
<li>Srivatsan, S., Famulok, M.. Functional Nucleic Acids in High Throughput Screening and Drug Discovery. Combinatorial Chemistry & High Throughput Screening. 2007;10(8), 698-705. https://doi.org/10.2174/138620707782507359</li>
<li>Labute. A Probabilistic Approach to High Throughput Drug Discovery. Combinatorial Chemistry & High Throughput Screening. 2002;5(2). https://doi.org/10.2174/1386207024607329</li>
<li>McMurray, C.. How organ-on-a-chip models can help drug development. Spectrum. 2023. https://doi.org/10.53053/thco8763</li>
<li>Kim, J., Lee, H., Selimović, Š., Gauvin, R., Bae, H.. Organ-On-A-Chip: Development and Clinical Prospects Toward Toxicity Assessment with an Emphasis on Bone Marrow. Drug Safety. 2015;38(5), 409-418. https://doi.org/10.1007/s40264-015-0284-x</li>
<li>Soragni, C., Rabussier, G., de Windt, L. J., Trietsch, S. J., Lanz, H. L., Ng, C. P.. High throughput assay to quantify oxidative stress in organ-on-a-chip placenta models in a multi-chip platform. Placenta. 2021;112, e26. https://doi.org/10.1016/j.placenta.2021.07.087</li>
<li>Vulto, P.. Organ-on-a-Chip models in early-stage drug discovery: A phenotypic screening exercise. Proceedings for Annual Meeting of The Japanese Pharmacological Society. 2022;96(0), 2-B-S20-4. https://doi.org/10.1254/jpssuppl.96.0_2-b-s20-4</li>
<li>Finkbeiner, S., Skibinski. Drug discovery in Parkinson's disease: update and developments in the use of cellular models. International Journal of High Throughput Screening. 2011, 15. https://doi.org/10.2147/ijhts.s8681</li>
<li>Upadhyaya, S., Selvaganapathy, P. R.. Microfluidic devices for cell based high throughput screening. Lab Chip. 2010;10(3), 341-348. https://doi.org/10.1039/b918291h</li>
<li>Paternoster, S., Nicolas, A., Schavemaker, F., Rabussier, G., Trietsch, S., Lanz, H.. LP-46 Organ-on-a-chip meets high throughput screening: measuring trans endothelial electrical resistance. Toxicology Letters. 2022;368, S300-S301. https://doi.org/10.1016/j.toxlet.2022.07.785</li>
<li>Chen, X., Oke, A., Sahin, D., Shang, Y.. High Throughput Screening for Drug Discovery and Virus Detection. Combinatorial Chemistry & High Throughput Screening. 2022;25(9), 1518-1533. https://doi.org/10.2174/1386207324666210811124856</li>
<li>Leung, C. M., Haan, P. d., Ronaldson-Bouchard, K., Kim, G., Ko, J., Rho, H. S.. A guide to the organ-on-a-chip. Nature Reviews Methods Primers. 2022;2(1). https://doi.org/10.1038/s43586-022-00118-6</li>
<li>Zhang, Y. S., Arneri, A., Bersini, S., Shin, S., Zhu, K., Goli-Malekabadi, Z.. Bioprinting 3D microfibrous scaffolds for engineering endothelialized myocardium and heart-on-a-chip. Biomaterials. 2016;110, 45-59. https://doi.org/10.1016/j.biomaterials.2016.09.003</li>
<li>Zhao, Z., Chen, X., Dowbaj, A. M., Sljukic, A., Bratlie, K. M., Lin, L.. Organoids. Nature Reviews Methods Primers. 2022;2(1). https://doi.org/10.1038/s43586-022-00174-y</li>
<li>Zhang, Y. S., Aleman, J., Shin, S. R., Kiliç, T., Kim, D. J., Shaegh, S. A. M.. Multisensor-integrated organs-on-chips platform for automated and continual in situ monitoring of organoid behaviors. Proceedings of the National Academy of Sciences. 2017;114(12), E2293-E2302. https://doi.org/10.1073/pnas.1612906114</li>
<li>Bhise, N. S., Manoharan, V., Massa, S., Tamayol, A., Ghaderi, M., Miscuglio, M.. A liver-on-a-chip platform with bioprinted hepatic spheroids. Biofabrication. 2016;8(1), 014101-014101. https://doi.org/10.1088/1758-5090/8/1/014101</li>
<li>Skardal, A., Murphy, S. V., Devarasetty, M., Mead, I., Kang, H., Seol, Y.. Multi-tissue interactions in an integrated three-tissue organ-on-a-chip platform. Scientific Reports. 2017;7(1). https://doi.org/10.1038/s41598-017-08879-x</li>
<li>Vinci, M., Gowan, S., Boxall, F., Patterson, L., Zimmermann, M., Court, W.. Advances in establishment and analysis of three-dimensional tumor spheroid-based functional assays for target validation and drug evaluation. BMC Biology. 2012;10(1), 29-29. https://doi.org/10.1186/1741-7007-10-29</li>
<li>Ramsay, R. R., Popović-Nikolić, M. R., Nikolić, K., Uliassi, E., Bolognesi, M. L.. A perspective on multi‐target drug discovery and design for complex diseases. Clinical and Translational Medicine. 2018;7(1), 3-3. https://doi.org/10.1186/s40169-017-0181-2</li>
<li>Mendes, B. B., Conniot, J., Avital, A., Yao, D., Jiang, X., Zhou, X.. Nanodelivery of nucleic acids. Nature Reviews Methods Primers. 2022;2(1). https://doi.org/10.1038/s43586-022-00104-y</li>
</ol>
</article>