Katrin Hoeck
Katrin Hoeck, Associate Director, Marketing and Business Development, Discovery Solutions, Lonza.

Clinical trial success rates remain low, with less than 10% of new drug compounds that enter Phase I studies ever making it to the market.1 To reduce attrition rates and improve productivity, many pharmaceutical researchers are re-evaluating their research and development (R&D) approach. With safety and efficacy being cited as the core factors responsible for the 90% failure rate, making the right choices during the early stages of the drug discovery process is key.

Start strong by choosing the right biological model

In 2018, AstraZeneca published an evaluation of its R&D success, stemming from its strategy to prioritize quality over quantity. The company introduced the now well-known “5R framework,” which guides early-stage decision-making by focusing on five key determinants of success: right target, right tissue, right safety, right patient, and right commercial potential.2

When the “5R framework” is used, it is possible to boost drug development success rates by making the right choices around target validation, hit and lead optimization, pharmacokinetic/pharmacodynamic modeling, and drug safety testing. Overall, this process helps improve the quality of candidate drug nomination before clinical trials begin.

One of the key considerations of the “5R framework” is using in vitro biological model systems that can best reflect the biology and mechanisms of a given disease. These models often provide predictions of drug efficacy and safety in humans that are more accurate than those provided by animal models. Consequently, when these models are used throughout the course of development, they can improve the chances of identifying a successful drug candidate.

Many of these models include living cells derived from tumors and tissues placed into culture conditions designed to mimic disease states and mechanisms. However, a major challenge in creating the right model is choosing the right cells to form its basis, with any system being only as good as the cells that it is created with.

Human primary cells are becoming the preferred method for building such physiologically relevant in vitro cell model systems. Primary cells that have been carefully isolated from human donor blood and tissue closely replicate the functions and mechanisms of the tissues from which they are derived.

Effectively isolating and purifying primary cells, however, is a challenging task, one requiring significant expertise in cell and tissue biology. Given the complexities, many laboratories could benefit from working with vendors to secure primary cells that have already been characterized, to ensure more confidence in building the right model for their needs. Once isolated, cells need to be validated for viability and functionality and be free of common laboratory pathogens. Tissues must also be ethically sourced from donors with the appropriate paperwork and authorization. Doing so can be a difficult and resource-intensive process.

Donor-specific characteristics are also an important consideration when building physiologically relevant models. For example, cytochrome P450 (CYP450) enzymes play a large role in determining the kinetics of distribution and clearance of small-molecule drugs throughout the body. The diversity of these enzymes is driven by both genetic and nongenetic factors, such as age, sex, and chronic diseases. This means that it is essential to understand the impact of CYP450 diversity in the metabolism of any small-molecule drug to effectively predict drug metabolism and pharmacokinetics in individual patients.

Ultimately, the determining factors for selecting the right in vitro biological model system are robustness and suitability for exploring the overarching research question.

Shifting trends and perspectives can drive drug discovery success

Advances in gene editing and 3D cell culture are transforming the way in vitro biological models are used to perform target validation, lead optimization, and preclinical safety testing.

More effective target validation and lead optimization: Effective target validation is key to any drug development program. Without a model system that accurately recapitulates in vivo mechanisms and phenotypes, there is significant risk of accepting a result that is not predictive of success in later studies. Using tools such as CRISPR-Cas9, even in hard-to-transfect human primary cells, enables researchers to validate targets in healthy and diseased cells by determining which genes are involved in disease development.3 The importance of using nonviral methods for transfection here cannot be overstated, as viral methods have been shown to raise the risk of off-target effects.4

In lead optimization, it is crucial that models accurately predict in vivo pharmacokinetics and pharmacodynamics. Such lead optimization studies are often performed in primary cells instead of cell lines to verify efficacy in a more predictive model. This reflects the industry push to shift attrition to the earliest possible stages of drug development, by evaluating efficacy and safety in tandem.

Improved predictivity in preclinical phases: There is a tremendous need for advanced, physiologically relevant human in vitro models for safety testing in place of animal models, particularly for many biotherapeutics where a suitable animal model is unavailable. To this end, the co-culture of human leukocyte antigen (HLA)-matched primary and immune cells is increasingly used to assess the immunogenicity of biotherapeutics. Additionally, with small-molecule safety, animal models are not a good match for human pharmacokinetics and continue to poorly predict safety in humans.

Human cell–based 3D cellular models and organ-on-a-chip microfluidic systems are increasingly being used to help characterize the safety of lead drug compounds, predicting how the drug would react with an organ in vivo. Although assay robustness and model complexity tend to hinder progress toward more widespread adoption, there is significant potential for these complex models to improve predictivity, enabling more confident decision-making around drug candidates.

Taking an integrated approach: Determining the optimal in vitro biological model for a given application can be daunting and often involves significant trial and error. However, adopting an integrated approach, where human primary cells, specialty media, and transfection protocols are optimized to work together, can save time and yield more consistent and reproducible results.

Cell-specific media formulations can optimize performance indicators for a more successful model system and provide ease of use and reproducibility. Specialty media can also aid the development of complex co-culture systems, whereas cell-specific, preoptimized nonviral transfection protocols remove the need for long upfront optimization cycles and ensure transfection success in primary cells. By using these complementary technologies, developers of in vitro biological model systems can better build application-specific models to drive drug discovery success.

Driving greater adoption of in vitro biological models

Regardless of their limited predictivity, animal models remain the preferred approach in drug safety studies. To date, there have been several instances where drugs that failed in animal models showed promising results when tested in vitro, that is, in human cell models. Yet, in vitro testing still has not seen widespread acceptance.

For the pharmaceutical industry to confidently adopt advanced in vitro model systems, there needs to be more adequate demonstration of their predictive capabilities. Correlation of animal in vitro and in vivo models should help to support the predictivity of human in vitro models. Currently, when a new drug compound fails due to efficacy or safety complications, this knowledge remains concealed within the records of the company performing the research. If more transparency were to be established across the industry, data cross-sharing would become possible, enabling companies to use the failed compounds to test the predictivity of human in vitro model systems and uncover the reasons behind the failure.


Choosing the right in vitro biological model lays a strong foundation for the entire R&D project, and combined learnings from advanced in vitro and in silico approaches during early drug discovery stages can improve clinical success rates. Rather than use the conventional “bench to bedside” approach, researchers can start from the bedside, where patient characteristics, tissue type, and physiological targets are well established. Doing so can help researchers reverse engineer the drug development pipeline and enable better-informed decision-making about which biological models to use. By taking an integrated approach to building in vitro biological model systems, researchers can also achieve more reliable results while saving time and lowering costs.

1. Dowden H, Munro J. Trends in clinical success rates and therapeutic focus. Nat. Rev. Drug Discov. 2019; 18(7): 495–496. DOI: 10.1038/d41573-019-00074-z.
2. Morgan P, Brown D, Lennard S, et al. Impact of a five-dimensional framework on R&D productivity at AstraZeneca. Nat. Rev. Drug Discov. 2018; 17(3): 167–181. DOI: 10.1038/nrd.2017.244.
3. Martufi M, Good RB, Rapiteanu R, et al.
Single-Step, High-Efficiency CRISPR-Cas9 Genome Editing in Primary Human Disease-Derived
Fibroblasts. CRISPR J. 2019; 2(1): 31–40. DOI: 10.1089/crispr.2018.0047.
4. Schober K, Müller TR, Gökmen F, et al. Orthotopic replacement of T-cell receptor α- and β-chains
with preservation of near-physiological T-cell
function. Nat. Biomed. Eng. 2019; 3(12): 974–984. DOI: 10.1038/s41551-019-0409-0.

Previous articleGaining New Insight on Receptor Tyrosine Kinase Signaling and Pharmacology
Next articleImmune Profiling System Redirected in Light of COVID-19