Biosimulation in Depth: A Detailed Guide

Biosimulation, which uses computer simulations of biological processes to predict the behavior of biological systems, is on a significant growth trajectory – thanks in part to the FDA’s strong recommendation for adopting biosimulation, an overall increase in predictive biosimulation in the research and development (R&D) process, and the recent use of biosimulation platforms for the development of COVID-19 vaccines. Now, as pharmaceutical and biotech companies continue to invest in artificial intelligence (AI)-enabled tools and technologies, we should expect to see even more widespread adoption of biosimulation across drug discovery and development. Here’s what you need to know.

What is biosimulation?‍

Biosimulation uses computer-based mathematical simulations to replicate biological processes, dynamics and systems. Model-based predictions like these allow researchers to gather important data about how biological systems behave without having to conduct such tests in living organisms, including humans and animals.

Also sometimes referred to as modeling and simulation (M&S), biosimulation models can be used in a range of different ways to help determine drug safety and efficacy, including drug identification, human dosing measures for different populations, drug-drug interactions (DDI), clinical trial design, and comparative effectiveness between drug candidates. To do this, biosimulation models use AI and machine learning (ML) to find patterns and evaluate relationships between drugs, patient populations and clinical trial parameters. 

Biosimulation technology is already making a huge impact in drug development and discovery, as pharma and biotech companies, contract research organizations (CROs), regulatory authorities, research institutions, and others turn to biosimulation software and services to help advance drug safety and efficacy, lower R&D costs, accelerate time-to-market, and advance our overall understanding of complex biological processes.

How is biosimulation used in drug R&D?‍

Pre-clinical testing of novel medicines, lead identification and optimization, and target identification and validation are some primary areas where biosimulation is being used in drug R&D. By applying principles from biology, chemistry, and pharmacology with proprietary algorithms, biosimulation forms the backbone of “virtual” drug trials, whose aim is to accurately predict the way medicines and diseases behave in the body. This work also includes gathering critical insights concerning drug interactions, safe dosing, toxicity predictions and more.

Biosimulation software in drug R&D typically covers:

• Molecular modeling and simulation
• Pharmacokinetic/pharmacodynamic (PK/PD) modeling and simulation
• Toxicity prediction software
• Physiologically based pharmacokinetic modeling and simulation (PBPK) 
• Trial design software

As one example of a recent application of AI-driven biosimulation in Drug R&D, VeriSIM Life used its BIOiSIM™ platform to find better therapeutic approaches in treating metabolic syndrome. In this instance, in silico development narrowed down two different drug classes with distinct modes of action and modalities. Generally, in silico modeling combines the advantages of both in vivo (in living organisms) and in vitro (in a test tube) experiments, but removes the ethical concerns and unpredictably more often associated with in vivo experimentation. Pharmacokinetic and pharmacodynamic profiles of the most promising drugs were modeled, showing predicted outcomes of combinatorial therapeutic interventions. 

How big is the biosimulation market?

Currently, the global biosimulation market (consisting primarily of sales of software and services) is being driven by considerable investments in R&D activities by pharmaceutical and biotechnology companies. Recent projections expect the biosimulation market to grow from US$3.17 billion in 2021 to US$3.64 billion in 2022, representing a compound annual growth rate (CAGR) of 15.0%. The global biosimulation market is expected to reach US$ ~10.3 billion by 2031.

The global biosimulation market has five major regions: North America, Europe, Asia Pacific, Latin America, and the Middle East & Africa. Currently, North America is currently the largest region in the biosimulation market due in large part to its reimbursement framework and its sophisticated healthcare infrastructure. Asia Pacific is showing the fastest growth, thanks to more plentiful CROs, a significant rise in healthcare IT spending, and its expanding healthcare infrastructure.

Biosimulation & personalized medicine

Today we are seeing a major shift from the conventional, one-side-fits-all approach to medical treatment to personalized, or precision, medicine, which uses genetic or other biomarkers to make tailored, patient-centric treatment decisions that consider differences in people's genes, environments, and lifestyles. This data-driven approach to patient stratification doesn’t replace clinician decision-making, diagnosis and prognostication, but augments it with AI-driven technologies like biosimulation which can be used to help uncover relationships between indications and biomarkers. Computation and inference modeling not only helps make better predictions around disease and treatment outcomes for individuals, it also helps find more diverse candidates for drug trials. [Further reading: Integrating AI/ML Models for Patient Stratification Leveraging Omics Dataset and Clinical Biomarkers from COVID-19 Patients: A Promising Approach to Personalized Medicine]

Some other advantages of AI-powered biosimulation

Artificial intelligence (AI) and machine learning (ML) help de-risk R&D decisions by providing meaningful actionable insights much earlier in the drug development process with unprecedented accuracy and scalability (for example, processing trillions of interdependencies and biosimulation scenarios to understand multiple biological systems at once). 

We see this advantage at work when it comes to animal testing, where AI-powered, digitally-rendered animal simulations are circumventing the need for animal testing in drug R&D. New drug compounds are often tested in animals for efficacy and safety before being used in human clinical trials – some countries even require animal studies as a condition of beginning clinical human trials. But in addition to raising serious ethical concerns, animal trials are notoriously slow and costly especially when it comes to large animals such as monkeys, with a staggeringly low success rate (about 90% of drug candidates tested in animals fail to make it to the drug pipeline). In AI-powered biosimulation, computer models help increase translatability and, in some cases, eliminate the need for unnecessary animal trials.

Biosimulation and ROI

The high failure rate of clinical trials in drug R&D is costly, wasteful, and hasn’t been addressed over the past few decades. But a comprehensive biosimulation platform can help de-risk critical decisions, reduce R&D waste across the board, lowering the expense of R&D and ultimately boosting ROI. AI-driven modeling and simulation helps maximize the commercial potential of new medications in a number of ways – optimizing dosages early, simulating specific patient populations, minimizing legal risks and so forth. Biosimulation can also improve cost efficiencies across the drug discovery and development process by predicting value drivers like efficacy, translatability, toxicity, adverse reactions and DDI at earlier stages of product development. 

What’s next for biosimulation?

As the need to increase efficiencies in drug development continues to drive market growth, AI-driven models and simulations will be increasingly tapped as a way to assess disease, evaluate cellular processes, and predict clinical outcomes. Demand for skilled professionals, particularly data scientists, will also intensify, with the competition for AI talent tightening across sectors and influencing adoption rates. And as health research tech continues to evolve, we should expect to see a continued shift away from traditional, statistical modeling to algorithm modeling and the growing adoption of AI-driven computer simulations in all phases of pharmaceutical R&D.