PHARMA – Twin – AXIOLOGO.ai

In a strategic collaboration, Axiologo and Novartis embarked on a transformative project named “Modelling dissolution curves and in vivo results,” also known as the Digital Twin of the drug. This initiative aimed to revolutionize the R&D process by predicting experimental outcomes, thus significantly reducing the time and resources required to bring new drugs to market.

Project Overview

Objective:

The primary goal was to predict the results of R&D experiments (batches) accurately, aiming to eliminate unnecessary experimental runs. This predictive approach was designed to expedite the R&D timeline and accelerate the drug’s entry into the market.

Challenge:

The project sought to address the inefficiencies in traditional pharmaceutical R&D, where multiple experimental runs escalate costs and delay market introduction. By harnessing AI technologies, the initiative aimed to streamline the development process.

Technologies Deployed:

Modelling: Python was used for creating predictive models.
Interface: Dash provided a user-friendly front-end for interacting with the models.

Execution and Results

Complexity in Prediction: Identifying key variables impacting the dissolution or in vivo curves and teaching the AI models to predict outcomes with less than 5% error margin represented the project’s core challenge.

Impact: Successfully eliminating one experimental run translated into a 15-25% reduction in time to market, saving up to €50,000 in process/material costs and approximately €40,000 in wages. Moreover, being the first generic drug on the market grants a six-month exclusivity period, yielding significant financial benefits.

Conclusions and Future Directions

Market Speed vs. Capital Savings: The project underscored the paramount importance of reducing time to market over mere cost savings.
Precision in Modelling: After just one or two experimental runs, the dissolution curve modelling achieved remarkable accuracy, underscoring the efficacy of the digital twin concept.
Insights into Design Space: The project provided invaluable insights into the variables affecting drug dissolution and in vivo results, enhancing future R&D efficiency.