How Temple Health plans to use digital twins to advance ALS care

Clinicians and researchers at Temple Health are working together on the Digital Twin for Personalized Medicine (DT4PM) project.to advance care for amyotrophic lateral sclerosis (ALS). By creating digital representations of people living with ALS, care teams are aiming to better predict disease progression, improve clinical trial design and empower patients with more personalized information to support decisions about their treatment and lifestyle priorities.  

Terry Heiman-Patterson, MD, professor of neurology at the Lewis Katz School of Medicine at Temple University and director of the Temple MDA/ALS Center of Hope, and Huanmei Wu, PhD, chair of the Department of Health Services Administration and Policy at Temple University’s College of Public Health, are leading the DT4PM project. They spoke with DHI about using AI and machine learning to create digital twins and how this framework could be a powerful tool for ALS treatment. 

Creating digital twins 

More than 40 different genes are associated with the risk of developing ALS. When people do develop the disease, there is significant variability across when and how it presents. Heiman-Patterson has had patients in high school and patients in their 90s. In some people, ALS affects speech and swallowing first. In others, it starts with their arms or legs.  

“We have this heterogeneity in terms of survival, in terms of progression, in terms of what parts of the motor system are most involved,” Heiman-Patterson said.  

Despite that heterogeneity, large quantities of ALS patient data amassed from various sources, such as the ALS/MND Natural History Consortium and ALS Knowledge Portal, offer a starting point for predicting disease progression in a patient.  

“We’ve already used some of that data for what we call landmark analysis to predict when they’re going to need a ventilator, to predict when they’re going to need a feeding tube…but it isn’t the same as a digital twin yet,” Heiman-Patterson said. 

Through the project, researchers are drawing on these large data sets and using AI methods with the goal of creating a framework to model the disease and simulate its progression in patients’ digital twins.  

“This digital twin is a dynamic representation of the patient’s health status,” Wu told DHI. “Our AI and machine learning is trying to identify what [are] significant factors which can affect their progress.” 

Modeling disease progression  

Right now, there is enough data that clinicians can start to model disease progression based on factors like age of onset and gender. Digital twins could give clinicians the ability to model disease progression on an individual level. 

“I can take an individual, and I can then model his disease and say: I predict you’re going to need a feeding tube in six months. I predict you’re going to need a ventilator in 10 months. I predict you’re going to live for 10 years,” Heiman-Patterson said.  

If used in clinical settings, care teams could use patient data collected on an ongoing basis to update their digital twins. As time goes on, they could simulate potential outcomes via predictive risk models applied to patients’ digital twins.  

“What if they pick a certain intervention, like a suggested treatment or medication?” Wu said. “The digital twin technology [can] show them: What’s the health benefit within three months or within six months?” 

Improving clinical trials for ALS treatment  

In addition to supporting the care of individual patients, digital twins could also impact the way clinical trials for ALS treatments are run. Currently, trials require a test group and a placebo group. Given the patient entry criteria for clinical trials, trial length and average life expectancy, most people living with ALS will only be able to participate in one trial, according to Heiman-Patterson.  

“Imagine how heartbreaking it is to find out you were on placebo for the only trial that you could be in,” she said.  

Digital twins have the potential to supplement clinical trial placebo groups, reducing the number of people taking placebos and enabling the evaluation of more drugs.  

For the patients taking drugs during clinical trials, digital twins could be a useful tool for predicting the efficacy of those drugs. For example, digital twins could incorporate biomarkers, like neurofilament light chain (NfL), which indicates nerve damage. The FDA approved the drug tofersen for patients with SOD1-ALS based on the drug’s ability to reduce NfL.  

“When we integrate things like biomarkers, like NfL, into our digital twin paradigm, it may even enable us to predict… whether or not a drug will be helpful,” Heiman-Patterson said. 

Building a future of more personalized ALS care 

Digital twins are an exciting prospect for ALS care, but Temple Health’s framework is still in development. Heiman-Patterson anticipates it will take two to five years to create these digital twins, depending on funding. The framework will also need to be validated. 

“With funding…we can collect more dynamic data and recruit a patient so we can verify if the digital twin can really work or not,” Wu said.  

If digital twins are integrated into clinical settings, Heiman-Patterson is hopeful that ALS patients will be able to make informed decisions about their care and better prioritize their time. 

“I can recommend a breathing machine or a feeding tube. The person still has the autonomy to accept or not accept those recommendations,” she said. “They need to maintain control over their decisions, but I’ll give them information that they can use to make an informed decision.” 

Digital twins have significant potential in more than just ALS care. This type of framework could be leveraged for the management and treatment of many other chronic diseases. But to get there, Wu emphasized the importance of collaboration across not only the research and clinical communities. She added, “We need to talk to the policymakers and insurance peers to get their buy-in in the long-term.” 

Carrie Pallardy, a Chicago-based freelance writer and editor, began her career covering healthcare more than a decade ago. Her work has taken into many different industries, but covering healthcare delivery remains a constant focus. She can be reached at [email protected] or on LinkedIn.