Introduction and Background of the Project

Introduction

Dengue and Zika are flavivirus infections prevalent in sub-Saharan Africa, South America, and Southeast Asia. Climate change, including warmer temperatures and changes in rainfall patterns, is increasing the geographical spread and scale of outbreaks of these mosquito-borne diseases. An estimated 3.9 billion people are already at risk of dengue infection. At present, there is no effective antiviral treatment for flavivirus infections. New medical tools are urgently needed to prevent severe illness and control outbreaks. In our previous screening project supported by the GHIT-Fund, we discovered several compounds that reduce viral replication in mammalian cells by blocking the viral proteins needed for replication. The current project aims to further improve these compounds with a view to developing new drug candidates for the treatment of flavivirus infections.

Project objective

The goal of this project is to develop lead compounds that can reduce replication of dengue and Zika viruses using hit compounds identified in the earlier screening project as a starting point. We aim to improve the antiviral activity of the hit compounds and refine important drug‑like properties such as solubility, lipophilicity, metabolic stability, and safety. By the end of the project, we aim to have identified compounds that lower the viral load in pre-clinical models and show enough promise to move forward to the next stage of drug development.

Project design

To improve the efficiency of compound optimization, we will use an AI‑supported medicinal chemistry approach that enables rapid and iterative cycles of compound design, synthesis, evaluation, and AI model refinement. In each cycle, biochemical and cell‑based antiviral activities, and data on solubility, lipophilicity, and metabolic stability will be used to guide the next round of compound design. As we accumulate more data, AI modelling will improve accuracy and support the identification of more promising compounds to synthesize. In addition, Drugs for Neglected Diseases initiative (DNDi) will perform advanced antiviral evaluations and early safety assessments, while Japan Institute for Health Security (JIHS) will evaluate compound efficacy against different viral strains and host cell-lines. These studies will help us assess the consistency and reliability of the efficacy of each compound as potential treatments for flavivirus infection. Promising compounds will be further evaluated in pre-clinical models, and these studies will assist in identifying lead compounds for further development.

How can your partnership (project) address global health challenges?

Flavivirus infections have serious health and economic impacts, especially in low‑ and middle‑income countries where healthcare resources are limited. Although vaccines exist, their availability and use remain limited, and many communities still lack practical protection against these diseases. Without effective antiviral drugs, outbreaks can spread quickly and become major public health emergencies. This project aims to deliver new antiviral drug candidates that can prevent disease progression, reduce severe cases, and strengthen preparedness for future flavivirus outbreaks. We are focused on developing small‑molecule medicines, which can be produced at relatively low cost, manufactured at scale, and distributed easily. These features make them well‑suited for regions with limited resources and may help reduce global health inequalities.

What sort of innovation are you bringing in your project?

This project aims to develop a new, innovative type of antiviral drug for flavivirus infections where no approved therapies currently exist. By pursuing a mechanism that differs from existing antiviral approaches, this study offers an opportunity to expand future treatment options and build a stronger foundation for responding to drug resistance and emerging viral threats. This project uses an innovative AI-supported approach to efficiently explore the chemical space where research resources are limited. By combining ligand‑based predictive AI models with generative AI models, we can improve both antiviral activity and drug‑like properties in parallel, potentially identifying strong lead compounds with fewer synthesized compounds than traditional medicinal chemistry requires. This efficient AI‑supported approach is valuable in the field of tropical diseases and may serve as a model for future drug discovery efforts in this area.

Role and Responsibility of Each Partner

Eisai leads the project and manages compound optimization, primary screening, evaluation of physical and stability properties, and integration of all data into the AI platform. DNDi performs advanced antiviral evaluations, safety assessments to support lead compound selection. JIHS evaluates antiviral activity across various viral strains and cell lines while also examining the mechanistic aspects of compound action. All three partners will participate in regular strategic meetings to share data, discuss the project direction, and make timely decisions regarding compound progression and issue resolution, ensuring that the project moves forward smoothly and effectively.