Back

Do Existing COVID-19 Vaccines Need to Be Updated in 2025?

Mellis, I. A.; Wu, M.; Wang, Q.; Bowen, A.; Gherasim, C.; Pierce, V. M.; Shah, J. G.; Purpura, L. J.; Yin, M. T.; Gordon, A.; Guo, Y.; Ho, D. D.

2025-05-05 microbiology
10.1101/2025.05.02.651777 bioRxiv
Show abstract

COVID-19 vaccines have been updated each year since 2022 to improve protection against evolving SARS-CoV-2 variants. However, it is unclear whether a reformulation will be necessary for 2025. KP.2-based monovalent COVID-19 mRNA vaccines (KP.2 MV) were authorized for use in 2024, and they conferred substantial protection against hospitalizations caused by viral variants that emerged and dominated later, such as KP.3.1.1 and XEC. Today, LP.8.1 and its subvariant LP.8.1.1 have become dominant worldwide, particularly so in North America. Other variants, such as the LF.7 subvariant LF.7.2.1, have emerged with a growth advantage in Asia. To characterize the antigenicity of LP.8.1, LP.8.1.1, LF.7, LF.7.2.1, and another variant under monitoring, MC.10.1, we tested serum samples from 20 individuals who recently received KP.2 MV in neutralization assays against JN.1, KP.2, KP.3, KP.3.1.1, XEC, LP.8.1, LP.8.1.1, LF.7, LF.7.2.1, or MC.10.1 pseudoviruses. Serum neutralizing antibody titers against LP.8.1, LP.8.1.1, LF.7, LF.7.2.1, and MC.10.1 were comparable to those against KP.3.1.1 and XEC, indicating that LP.8.1.1 and other recently dominant subvariants are antigenically similar to their predecessors. Therefore, the currently authorized KP.2 MV may not need to be updated for 2025, if the vaccine manufacturers could demonstrate comparable immunogenicity for KP.2 MV and LP.8.1-based mRNA vaccines and, of course, in the absence of an antigenically divergent SARS-CoV-2 variant emerging.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.