Mathematics and computer science are revolutionizing the way new drugs and treatments are tested and implemented. A new paper published in Chaos and written by U4 McGill Physiology and Math major Sofia Alfonso, postdoctoral researcher Adrianne L. Jenner, and Dr. Morgan Craig from the University of Montreal’s department of Math and Statistics, explores new alternatives to the challenges of using quantitative tools.

In the pharmaceutical industry, pre-clinical and clinical trials are challenging, time-consuming, and costly. Virtual drug trials allow researchers to better understand and treat complex diseases such as cancer, diabetes and depression. Conducted faster than conventional clinical trials, virtual drug trials also allow more efficient and affordable distribution of treatments to the public due to high recruitment rates, better compliance, and lower drop-out rates. The paper presents multiple case studies that test experimental medications using mathematical modelling. 

Such studies are considered in silico, as they are conducted by a computer program and do not involve live patients, as an in vivo trial would. These simulations can predict the effect of a medication on virtual patients, leading to important insights about a drug’s efficacy before investing time and money into human subject testing. 

Alfonso and Craig say that the pharmaceutical industry is already using in silico models for research and emphasize the need for collaboration between experimentalists and clinicians in order to develop more accurate and effective models.

“In drug development, for example, study of a novel drug delivery device for anti-HIV therapy contributed to its continued development and ongoing clinical trials of similar devices,” Craig wrote in an email to The McGill Tribune.

One case study explored the potential of mathematical modeling in the development of treatments for infectious diseases, such as the Herpes Simplex virus (HSV) or the Human Immunodeficiency Virus (HIV). Based on data of viral shedding collected from real patients, an experimental drug was administered to a virtual patient infected with HSV. Researchers then optimized the drug’s dose for clinical trials, paving the way for future studies of similar drugs for the antiviral treatment of HIV and HSV. 

“A big challenge is finding adequate parameters in the literature such that the model can be accurately calibrated,” Alfonso wrote in an email to the Tribune, referring to the need for data from clinicians to construct accurate models. “Thus, collaborative efforts that bridge quantitative approaches with experimental work can be integral to developing useful models.”

The researchers are optimistic about the potential of virtual trials in the development of treatments more quickly and less invasively during public health crises such as COVID-19. Transitioning to remote trials could limit the risks of in-person contact, especially in medical settings. 

“Currently, we have been working with an interdisciplinary team on modelling COVID-19 in virtual patients, allowing us to simulate the mechanisms resulting in severe SARS-CoV-2 infection,” Alfonso wrote. “I am hopeful that as we gain more data, our model will provide further clinically relevant findings.”

Given the importance of quantitative methods in physiology, Alfonso emphasizes the opportunities for future physiology students willing to delve into mathematics, physics, and computer science. 

Craig also calls upon physiology students to maintain an open mind on these disciplines, noting that quantitative methods are already being implemented by the industry.

“Many researchers have summer positions for undergraduates that provide hands-on training,” Craig wrote. “In fact, Sofia [Alfonso] started in my lab as a PHGY 461 student and has continued as a research assistant since.”