MADE: A Computational Tool for Predicting Vaccine Effectiveness for the Influenza A(H3N2) Virus Adapted to...

Area of Science:

• Virology
• Immunology
• Computational Biology

Background:

• Seasonal influenza H3N2 poses a significant public health risk, with current vaccine effectiveness (VE) being suboptimal.
• Viral passage in embryonated eggs is a critical step in influenza vaccine production.
• Egg passage adaptation, driven by convergent evolution in hemagglutinin (HA1), negatively impacts VE.

Purpose of the Study:

• To leverage the negative correlation between egg passage adaptation and VE.
• To develop a computational tool for selecting optimal candidate vaccine viruses (CVVs).

Main Methods:

• Utilized mutational mapping, a probabilistic approach, to analyze egg passage adaptation patterns.
• Developed a new metric, adaptive distance (AD), to quantify the strength of egg passage adaptation.
• Validated the correlation between AD and influenza H3N2 VE.

Main Results:

• Adaptive distance (AD) was found to be negatively correlated with influenza H3N2 vaccine effectiveness (VE).
• AD explained approximately 75% of the variability in VE.
• A computational package, MADE (Measure the Adaptive Distance and predict vaccine Effectiveness), was developed.

Conclusions:

• The MADE package offers a robust tool for assessing egg passage adaptation effects.
• It enables the selection of reliable influenza vaccine strains with minimal egg-passaged changes.
• This can lead to improved seasonal A/H3N2 influenza vaccine efficacy.