In This Issue
Winter Bridge on Frontiers of Engineering
December 15, 2023 Volume 53 Issue 4
This issue features articles by 2023 US Frontiers of Engineering symposium participants. The articles cover pressing global issues including resilience and security in the information ecosystem, engineered quantum systems, complex systems in the context of health care, and mining and mineral resource production.

Engineering Human Organoids and Organs-on-Chips for Disease Modeling, Drug Development, and Personalized Medicine

Wednesday, December 13, 2023

Author: Janny Piñeiro-Llanes and Rodrigo Cristofoletti

Human microphysiological systems have the capacity to vastly improve drug development by providing more physiologically relevant and predictive models for drug testing.

The pharmaceutical industry has been struggling to sustain sufficient innovation to replace the loss of revenues due to patent expirations for successful products and increasing R&D costs. A key aspect of this problem is the decreasing number of truly innovative new medicines approved by the US Food and Drug Administration (FDA) and other major regulatory bodies around the world (Paul et al. 2010). Among all the challenges the pharmaceutical industry faces, we argue that improving R&D productivity remains the most important. In fact, reducing the high attrition rates in drug development continues to be a key challenge for the pharmaceutical industry.

The Need to Improve Preclinical Translational Models to Decrease Attrition Rates in Drug Development

Despite the significant technological improvements in drug development in recent decades, attrition rates remain high. Although the number of failures of small-molecule drug candidates due to poor pharmacokinetic profiles diminished significantly in recent years, the failure rate is increasing because of ­­efficacy and safety issues, which keeps the overall attrition rates high. An analysis of combined data on the attrition of drug candidates from AstraZeneca, Eli Lilly and Company, ­GlaxoSmithKline, and Pfizer revealed that the primary causes of failure for terminated compounds in phases 1 and 2 were clinical safety issues and inadequate efficacy, respectively (Waring et al. 2015). These compounds’ failure despite the satisfactory results from preclinical animal studies highlights the need to improve preclinical studies.

Animal studies have been a fundamental component of pharmaceutical research for many decades. While animal models provide insights into systemic effects and long-term consequences that are challenging to replicate in vitro, concerns about the capacity of animal data to anticipate human responses have emerged among ­scientists in the field. Indeed, an unparalleled study assessing the correlation between outcomes of pre­clinical models and results seen in early-phase clinical trials in oncology reported that animal toxicity did not show a strong correlation with human toxicity, with a median positive predictive value of 0.65 (significant proportion of false positive results or sponsor risk) and a negative predictive value of 0.50 (significant proportion of false negative results or patient risk) (Atkins et al. 2020). Furthermore, different genomic responses between animal models, including genetically modified mice, and humans suggest that molecular results from current mouse models developed to mimic human diseases may fail to translate directly to human conditions (Seok et al. 2013). For example, the genetic basis of Down Syndrome in humans is related to the trisomy of chromosome 21, whereas in the genetically modified Ts65Dn mice, the triplicated segment consists of ∼104 HSA21 orthologs as well as 60 centromeric MMU17 genes that are not triplicated in humans with Down Syndrome (­Antonarakis et al. 2004). As a result, the prototypical compound RG1662 improved performance in cognitive tests in TS65Dn mice, but a subsequent phase 2 clinical trial was stopped prematurely due to the lack of efficacy in humans (Erazo-Oliveras et al. 2023; Rudolph and Möhler 2014). Altogether, these examples demonstrate that relying only on animal data for predicting human responses to new drugs is an inadequate preclinical strategy.

PL figure 1 - Copy.gif

New drugs in the development pipeline (e.g., ­monoclonal antibodies) are also becoming highly ­specific to human targets. Yet, the relevance of animal studies to assess drug efficacy and safety has been questioned, like in the case of a novel superagonist anti-CD28 ­monoclonal antibody (TGN1412) that directly stimulates T cells. In preclinical models, the stimulation of CD28 with TGN1412 ­resulted in transient lymphocytosis with no detectable toxic or pro-inflammatory effects. However, within ninety minutes after receiving a single intravenous dose of the drug, six healthy volunteers in a phase 1 study had a systemic inflammatory response characterized by a rapid induction of proinflammatory cytokines and accompanied by headache, myalgias, nausea, ­diarrhea, erythema, vaso­dilatation, and hypotension. Within twelve to sixteen hours after infusion, the healthy volunteers became critically ill, with pulmonary infiltrates, lung injury, renal failure, and disseminated intravascular coagulation (Suntharalingam et al. 2006). Besides the limited translatability of preclinical data, drug development is significantly affected by an unprecedented shortage of non-human primates for preclinical studies. A recent US National Academies of Sciences, Engineering, and ­Medicine report concluded that the situation is compromisingly critical for bio­medical research now—and will continue to be well into the future (NASEM 2023). Overall, the combination of limited translatability of preclinical data and a deficiency of non-human primates for preclinical studies urge the development of alternative approaches in preclinical research.

Relying only on animal data for predicting human responses to new drugs is an inadequate preclinical strategy.

Human Microphysiological Systems:
A Promising Alternative

In this context, developing human-relevant alternatives to animal testing could improve the translatability of nonclinical models. Human microphysiological systems (MPSs), such as organoids and microfluidic organs-on-chips, ­rapidly evolve as promising in vitro tools that can recapitulate human physiology by recreating key biological processes and disease states. MPSs combine ­microsystems engineering with cell biology, yielding cell-culture models that can display three-dimensional architecture, multi­cellular interactions, tissue-tissue interfaces, fluid flow, and organ-level mechanical cues (Roth 1979). The current working hypothesis in the field is that improving the physiological relevance and complexity of the in vitro system would improve its predictive capacity. The obvious caveat is that the more complex an in vitro model is, the less compatible it will be with the high throughput screening concept, so there is plenty of room for engineering innovation.

Initially, the adoption of MPSs was primarily for preclinical safety (drug toxicology and metabolism), often for applications in a defined context of use requiring limited validation efforts. However, the MPS concept gained traction among regulators and rapidly transitioned from academic curiosity to regulatory acceptance. Indeed, on December 29, 2022, the FDA Modernization Act 2.0 was signed into law. The bill essentially incorporates MPSs into the US legal framework. This bill marked a significant shift in the regulatory paradigm, moving from interspecies translation to inter-systems and in vitro-in vivo extrapolation. Developing an MPS is a multi­disciplinary effort that starts with the identification of the scientific question to be addressed and may encompass two ­parallel branches: 1) biology (e.g., cell source, bio­materials, scaffold), and 2) engineering (e.g., materials, micro­fabrication, sensors/actuators integration). Then, both branches are integrated, resulting in a microfluidic organ-chip device. The main feature of an MPS is its ability to recapitulate organ-level architecture and functionality, significantly differing from traditional cell-based assays. After functional validation of the MPS it can be applied to inform drug development and personalized medicine decisions (Rogal et al. 2022).

The main feature of an MPS is its ability to recapitulate organ-level architecture and functionality, significantly differing from traditional cell-based assays.

Organoids, which are self-organizing, 3D culture systems that are highly similar to—and in some cases, ­histologically indistinguishable from—actual human organs, are another type of MPS. One common feature of all organoids is that they are generated from pluripotent stem cells (PSCs) or adult stem cells by mimicking human development or organ regeneration in vitro.

The development of organoid technology is still in its infancy compared to established cell lines and animal models, with challenges still to be overcome. Never­theless, the prospect of organoids complementing existing model systems to extend basic biological and medical research and drug discovery into a more physiologically relevant human setting is becoming ever more widely appreciated (Kim et al. 2020). For example, FDA-approved drug screening in patient-derived ­organoids demonstrates the potential of drug repurposing for rare cystic fibrosis genotypes. Briefly, cystic fibrosis is a rare hereditary disease caused by mutations in the CFTR gene. Pharmaco­therapies termed CFTR modulators that rescue CFTR function have revolutionized treatment for approximately 85% of people with cystic fibrosis who carry the most prevalent F508del-CFTR mutation. Never­theless, a large unmet need remains to identify new and affordable treatments for patients with CFTR mutations that are non-eligible for or non-responsive to CFTR ­modulators. ­Consequently, a recent study leverages the fact that CFTR function measurements in patient-derived intestinal organoids are associated with clinical features of cystic fibrosis to test drug repurposing in a personalized setting using a high-throughput screening. These CFTR function measurements are performed by means of the forskolin-induced assay, in which forskolin induces fluid secretion into the organoid lumen, resulting in rapid organoid swelling in a (near-to) complete CFTR-­dependent ­manner (de Poel et al. 2023).

Our lab is highly interested in using MPSs to study genetic diseases and identify druggable targets. For example, understanding the molecular and cellular mechanisms downstream of the extra copy of chromosome 21 in Down Syndrome is one of our research interests. By using pluripotent stem cells from mosaic Down ­Syndrome individuals, we were able to develop in vitro twin organoids discordant for 21 trisomy, which would enable us to assess the perturbations of gene expression in trisomy 21 and to eliminate the noise of genomic variability. Additionally, we can derive matched organs for transcriptomics, ­proteomics, and pharmacology studies to shed some light on the source of multiple phenotypic variants among ­people living with Down Syndrome ­(figure 1).


Reducing drug attrition rates is of paramount importance in the pharmaceutical industry and has significant implications for both industry and public health. Specifically, improving efficiency, cost-effectiveness, and overall drug development success is critical for the pharmaceutical industry’s sustainability, innovation, and ability to provide patients with safe and effective treatments. Moreover, it has a direct impact on public health by ensuring that patients have timely access to new and improved medications. MPSs have the potential to revolutionize drug development by providing more physiologically relevant and predictive models for drug testing. By doing so, they can significantly reduce high attrition rates, leading to more efficient, cost-effective, and patient-centric drug development processes.


Antonarakis SE, Lyle R, Dermitzakis ET, Reymond A, Deutsch S. 2004. Chromosome 21 and Down syndrome: From genomics to pathophysiology. Nat Rev Genet 5(10):725–738.

Atkins JT, George GC, Hess K, Marcelo-Lewis KL, Yuan Y, Borthakur G, Khozin S, LoRusso P, Hong DS. 2020. Pre-clinical animal models are poor predictors of human toxicities in phase 1 oncology clinical trials. Br J Cancer 123(10):1496–1501.

de Poel E, Spelier S, Hagemeijer MC, van Mourik P, Suen SWF, Vonk AM, Brunsveld JE, Ithakisiou GN, Kruisselbrink E, Oppelaar H. 2023. FDA-approved drug screening in patient-derived organoids demonstrates potential of drug ­repurposing for rare cystic fibrosis genotypes. Journal of Cystic Fibrosis 22(3):548–559.

Erazo-Oliveras A, Muñoz-Vega M, Mlih M, Thiriveedi V, ­Salinas ML, Rivera-Rodríguez JM, Kim E, Wright R, Wang X, Landrock K, and five others. 2023. Mutant APC reshapes Wnt signaling plasma membrane nanodomains by altering ­cholesterol levels via oncogenic β-catenin. Nat Commun 14, 4342.

Kim J, Koo BK, Knoblich JA. 2020. Human organoids: model systems for human biology and medicine. Nat Rev Mol Cell Biol 21:571–584.

National Academies of Sciences, Engineering, and Medicine (NASEM). 2023. Nonhuman Primate Models in Biomedical Research: State of the Science and Future Needs. ­Washington, DC: The National Academies Press.

Paul SM, Mytelka DS, Dunwiddie CT, Persinger CS, Munos BH, Lindborg SR, Schacht AR.  2010. How to improve RD productivity: the pharmaceutical industry’s grand challenge. Nat Rev Drug Discov 9(3):203–214.

Rogal J, Schlünder K, Loskill P. 2022. Developer’s Guide to an Organ-on-Chip Model. ACS Biomater Sci Eng 8(11):4643–4647.

Roth A. 1979. Human microphysiological systems for drug development. Science 373(6561):1304-1306.

Rudolph U, Möhler H. 2014. GABAA receptor subtypes: Therapeutic potential in down syndrome, affective disorders, schizophrenia, and autism. Annu Rev Pharmacol Toxicol 54:483–507.

Seok J, Warren HS, Cuenca AG, Mindrinos MN, Baker HV, Xu W, Richards DR, McDonald-Smith GP, Gao H, Hennessy L, and twenty-nine others. 2013. Genomic responses in mouse models poorly mimic human inflammatory diseases. Proc Natl Acad Sci U S A 110:3507–3512.

Suntharalingam G, Perry MR, Ward S, Brett SJ, Castello-Cortes A, Brunner MD, Panoskaltsis N. 2006. Cytokine Storm in a Phase 1 Trial of the Anti-CD28 Monoclonal Antibody TGN1412. New England Journal of Medicine 355: 1018-1028.

Waring MJ, Arrowsmith J, Leach AR, Leeson PD, Mandrell S, Owen RM, Pairaudeau G, Pennie WD, Pickett SD, Wang J, and two others. 2015. An analysis of the attrition of drug candidates from four major pharmaceutical companies. Nat Rev Drug Discov 14(7):475–486.


About the Author:Janny Piñeiro-Llanes is postdoctoral researcher and Rodrigo Cristofoletti is assistant ­professor, both at the Center for Pharmacometrics and Systems Pharmacology, Department of Pharmaceutics, University of Florida.