The University of Massachusetts Amherst
University of Massachusetts Amherst

Search Google Appliance


UMass Amherst Engineer on International Research Team Developing New Tools for Reducing Drug Dosage Used to Treat HIV

Graph shows how AI-PRS is used to determine drug dosage.

Graph shows how AI-PRS is used to determine drug dosage

Seeking Minimum Drug Levels to Reduce Side Effects

AMHERST, Mass. – An international team of researchers from the University of Massachusetts Amherst, the University of California, Los Angeles (UCLA) and universities in China and Singapore, is using artificial intelligence to develop a system that minimizes the dosage of drugs used to treat human immunodeficiency virus (HIV) as a way to reduce side effects. The findings are published in the journal Advanced Therapeutics.

The team includes Tingyi “Leo” Liu, assistant professor of mechanical and industrial engineering at UMass Amherst. Liu is head of the Interdisciplinary Interface Engineering Laboratory (Inter²EngrLAB) in the College of Engineering.


Liu and the team say antiretroviral therapy (ART), using a combination of drugs, is a mainstay in treating HIV infection and making it a controllable chronic illness. He also says an HIV patient is traditionally administered the same ART regimen for life, even after his/her viral load has been reduced by several orders of magnitude from the initial viral load.


Because the drugs used to control the disease have side effects, dose reduction is an important goal once the level of HIV is reduced and patient health is stabilized. Traditional dosing regimens can result in side effects that range from mild to severe, including kidney damage. Studies have shown that more than half of the patients discontinued one or more of the drugs because of the side effects.Previously, dose reduction has only been explored in a trial‐and‐error manner, Liu says.


Using artificial intelligence (AI), the team has discovered that the drug and dose inputs can be linked to the output of viral load reduction through a simple mathematical relationship depicted as a parabolic response surface (PRS), Liu says.

He says the team chose to work with three drugs approved and recommended by the World Health Organization that are used in combination to treat HIV-positive individuals. He says the goal was to determine a way to provide patients with a combined dosage of the drugs that is sufficient to maintain their health but low enough to reduce side effects.


“We needed to find a method to lower the dose to a minimum amount that won’t allow the HIV virus to come back, but also reduces the side effects,” Liu says.


Using a small group of patients, the research team used the new AI-PRS platform to study the impact of lowering the dosage of one of the drugs by about a third and found it kept the virus load steady and below the detectable limit. For the first time, the minimum drug dose was determined based on a rational and clinically-actionable approach rather than the conventional trial-and-error strategies. The team believes this shows that the new method can serve as a model for the long-term management of HIV, as well as other infectious diseases.


Liu also says this new method for setting dosing may be further refined so that ART can be developed for individuals to have their personalized regimen for treatment.


Besides UMass Amherst and UCLA, the other members of the research team come from Fudan University and Shanghai Jiao Tong University in China, and the National University of Singapore.


In addition to his faculty appointment in the College of Engineering, Liu is affiliated with the Institute for Applied Life Sciences (IALS), which combines deep and interdisciplinary expertise from 29 departments on the UMass Amherst campus to translate fundamental research into innovations that benefit humankind.


November 25, 2019

Contact: Tingyi “Leo” Liu 413/577-0927