In recent years, automation and artificial intelligence (AI) have been huge trends within the clinical trials landscape, as companies look to incorporate such technologies to optimise their operational efficiency.
Alongside the burgeoning growth of AI in the sector, digital twin technology is emerging as a powerful tool – one that could help shape the future of clinical trials.
Go deeper with GlobalData
Discover B2B Marketing That Performs
Combine business intelligence and editorial excellence to reach engaged professionals across 36 leading media platforms.
This technology, which has its roots in the factory setting, could one day allow clinical trial operators to create a dynamic, virtual replica of a patient’s physiology. In theory, this could allow operators to predict how a patient might respond to a treatment by harnessing real-world data.
At the 2025 Outsourcing in Clinical Trials DACH conference in Zurich, during a panel entitled “The Future of Digital Twins in Clinical Development,” Dimitris Christodoulou, global business lead for digital health at Roche, observed that the ultimate aim for digital twins – fully simulating a patient’s physiology -remains “on the horizon.” He emphasised that his comments reflected his personal views and did not constitute an official corporate position.
The current potential of digital twins
While the digital twin concept has progressed since its first introduction into the clinical trials landscape, Christodoulou notes that the focus and priority for the technology now should be on delivering immediate value.
“The potential for immediate breakthroughs is more related to Operational and ‘behavioural’ twins,” he stated.
US Tariffs are shifting - will you react or anticipate?
Don’t let policy changes catch you off guard. Stay proactive with real-time data and expert analysis.
By GlobalData“Digital twins will take first ground in this context, as its use case delivers proven efficiency today. Then, the concept may be brought to patients and other scenarios,” Christodoulou added.
Currently, digital twins are being used in pilot and exploratory settings to optimise trial design, as they can simulate millions of scenarios to pinpoint where costs can be saved and time can be reduced.
These models can also be used to predict if an individual has a “high probability of dropping compliance or dropping off within the next 30 days”, according to Christodoulou.
To get the most out of a digital twin, Christodoulou believes that they will require input from a diverse range of sources, including everything from “multi-omics data, demographics and electronic health records to real-time data from wearables and recruitment channels”.
Following the incorporation of these efforts, Christodoulou suggest that the industry could see a robust twin or tool in the next five-to-seven years.
Digital twins yet to overcome hurdles
When employing digital twins in the clinical trials landscape, the industry’s lack of full mechanistic understanding of disease can present a challenge, as some information cannot be translated from the molecular to the organism level.
However, a hybrid approach could be the answer, as this allows companies to harness established mechanistic level information while bridging the gaps using AI.
“While you might not fully comprehend the reasoning behind the suggestions, they could still prove useful,” Christodoulou said.
Alongside issues at a molecular level, data can also present a hurdle for digital twin development. “There is still significant data fragmentation across the ecosystem, leading to the creation of models which end up using information that is simply not useful,” Christodoulou iterated.
Companies developing this technology will also have to overcome ethical and regulatory concerns, which, in part, centre around the use of historical data.
“It would be easy for these biases to creep into any digital twin solution, so there is still a lot to uncover to ensure fairness and equity,” Christodoulou commented.
Though there is still some way to go before digital twins reach their true end-game, Christodoulou believes that the industry is “approaching the horizon where we will see robust, scalable digital twin tools for clinical trial optimisation”.
