Virtual clinical trial populations in motor neuron disease
Lead Research Organisation:
King's College London
Department Name: Clinical Neuroscience
Abstract
This project aims to develop a set of digital twins of people with motor neuron disease. We will then carry out a clinical trial using computer simulations to replicate what would happen in real life. We can use that knowledge to improve the design of a real trial, making it shorter and more efficient. The digital twins will also mean that in the real trial, fewer people will need to take a blank version of the treatment (placebo) and more people will be able to take the real treatment, which is very important in a terminal condition like motor neuron disease.
Aim of the project
The aim of this project is to make a computer-generated population of people with MND. These virtual people, or "digital twins" can then be used to greatly improve clinical trials, reducing the number of people who need to take placebo in the real world, improving the design of studies, improving the rules for inclusion and exclusion, and ensuring that trial timelines are more likely to be met.
Why is this project needed?
Our understanding of what causes MND is rapidly improving, and as a result there are many new potential treatments that need to be tested in clinical trials. To be accepted as valid evidence for licensing a new therapy, a clinical trial needs to give some people a placebo, rather than the active drug. In a disease that is inevitably fatal, like MND, the use of placebo leads to ethical concerns, and the field has responded by shortening trials and randomizing more people to treatment arms than placebo arms. These changes make it more difficult to detect a treatment effect, and other approaches are needed. One option is to model MND to generate virtual populations that have the same characteristics as real trial participants.
What difference will this research make?
This modelling would allow more accurate prediction of the effects of different clinical trial designs, might be used to predict the potential results of a trial even if everyone were on active therapy (no placebo), and could be used to supplement real trial populations to reduce the need for large numbers of people on placebo.
What this project involves
This project will use various methods to generate virtual populations of people with motor neuron disease, comparing the performance of each one with real MND populations. The methods will include simulation, database manipulation, artificial intelligence and machine learning. Real clinical trial data will be used to test the behaviour of the virtual populations, assessing the strengths and drawbacks of each approach. The final method will use a combination of all successful techniques to produce a robust, high performing virtual MND population, suitable for trial design and analysis.
Sharing of results and how the work relates to other research
The project is complementary to other research in the UK and internationally. There are other groups working on virtual trial populations using alternative approaches, and we will share our methods and results to inform each other. The project will directly impact clinical trials research. All our results will be shared with the general public through social media (primarily X and LinkedIn) using language understandable by the general public, as well as through more traditional channels such as open access scientific publications.
Aim of the project
The aim of this project is to make a computer-generated population of people with MND. These virtual people, or "digital twins" can then be used to greatly improve clinical trials, reducing the number of people who need to take placebo in the real world, improving the design of studies, improving the rules for inclusion and exclusion, and ensuring that trial timelines are more likely to be met.
Why is this project needed?
Our understanding of what causes MND is rapidly improving, and as a result there are many new potential treatments that need to be tested in clinical trials. To be accepted as valid evidence for licensing a new therapy, a clinical trial needs to give some people a placebo, rather than the active drug. In a disease that is inevitably fatal, like MND, the use of placebo leads to ethical concerns, and the field has responded by shortening trials and randomizing more people to treatment arms than placebo arms. These changes make it more difficult to detect a treatment effect, and other approaches are needed. One option is to model MND to generate virtual populations that have the same characteristics as real trial participants.
What difference will this research make?
This modelling would allow more accurate prediction of the effects of different clinical trial designs, might be used to predict the potential results of a trial even if everyone were on active therapy (no placebo), and could be used to supplement real trial populations to reduce the need for large numbers of people on placebo.
What this project involves
This project will use various methods to generate virtual populations of people with motor neuron disease, comparing the performance of each one with real MND populations. The methods will include simulation, database manipulation, artificial intelligence and machine learning. Real clinical trial data will be used to test the behaviour of the virtual populations, assessing the strengths and drawbacks of each approach. The final method will use a combination of all successful techniques to produce a robust, high performing virtual MND population, suitable for trial design and analysis.
Sharing of results and how the work relates to other research
The project is complementary to other research in the UK and internationally. There are other groups working on virtual trial populations using alternative approaches, and we will share our methods and results to inform each other. The project will directly impact clinical trials research. All our results will be shared with the general public through social media (primarily X and LinkedIn) using language understandable by the general public, as well as through more traditional channels such as open access scientific publications.
