Implications of the apparent extinction of B/Yamagata-lineage human influenza viruses

An unusual and tangential outcome of the COVID-19 pandemic has been the apparent extinction of one of the four circulating varieties of human influenza viruses, known as the B/Yamagata/16/88-lineage (B/Yam) viruses^1,2. This lineage of influenza viruses has circulated globally since the mid-1970’s when influenza B viruses diverged into two genetically and antigenically distinct lineages, known as B/Victoria/2/87-lineage (B/Vic) and B/Yam³. B/Yam viruses predominated throughout the 1990s while B/Vic viruses were limited to circulation in China; however, B/Vic viruses re-emerged outside China in the early 2000’s⁴ and the two lineages then co-circulated globally, albeit in varying proportions from year to year^5,6 and from country to country; the figure (Fig. 1) shows the varying circulation patterns of B-lineages from January 2000 to July 2024 in Australia from samples submitted to the WHO Influenza Centre. Most recently in 2018, B/Yam viruses were the predominant B-lineage, accounting for approximately 90% of the world’s typed influenza B viruses (https://www.who.int/tools/flunet). Following the onset of the COVID-19 pandemic and the measures put in place to ameliorate its impact in early 2020, the number of influenza viruses circulating globally fell to historically low numbers. This trend continued throughout 2020 and most of 2021, with the number of influenza viruses reported to the WHO returning to normal in late 2021 and with seasonality and circulation patterns returning to normal in 2022 (https://www.who.int/tools/flunet).

Y (B/Yam) or V (B/Vic) at the top of the graph indicates which B-lineage was recommended by the WHO for inclusion in the Southern Hemisphere trivalent vaccine for that year.

Full size image

During this time, from March 2020 to November 2024, no B/Yam viruses have been confirmed worldwide. The last B/Yam virus reported to the sequence database GISAID (with haemagglutinin (HA) gene sequence) was B/Arizona/12/2020 (EPI_ISL_430780) collected on 24 March 2020 (https://gisaid.org/). While there have been reports of B/Yam viruses detected after this date, they have either been misreported, incorrectly assigned or have not been confirmed by sequencing. Hence, at this point in time, it appears that B/Yam viruses have been eliminated from human circulation due to the bottleneck that occurred during the COVID-19 pandemic⁷ leaving just B/Vic viruses in circulation (along with the two influenza A subtypes: influenza A(H3N2) and A(H1N1)pdm09).

What does this mean for human influenza vaccines and for the handling of B/Yam viruses in the laboratory? In the short term, the extinction of B/Yam viruses means that a B/Yam component is no longer required in the influenza vaccine and therefore the vaccine can now revert to a trivalent vaccine composed of just 3 viral components representing the circulating influenza virus types: A(H1N1)pdm09, A(H3N2) and B/Vic, instead of quadrivalent vaccines which also included a B/Yam component². Influenza vaccines with two influenza B components have only been used since 2013. Quadrivalent vaccines were introduced in order to broaden immunity against influenza B viruses as the proportions of the B/Yam:B/Vic each year could vary widely and the predominant lineage was difficult to predict at the time of vaccine virus selection⁵. Influenza vaccine manufacturers are now in the process of working with regulators to have trivalent influenza vaccines approved (or re-approved) for use in the upcoming influenza seasons. The US will be one of the first countries to switch from using quadrivalent vaccines to using only trivalent influenza vaccines for all vaccine platforms (live attenuated, inactivated, recombinant protein) for their 2024–5 influenza season. Astra Zeneca’s live attenuated influenza vaccine (LAIV) will revert to a trivalent vaccine in many countries in 2024–25, although some countries, such as Canada, will still have a quadrivalent LAIV for their 2024–25 season. Other countries/regions will introduce trivalent inactivated or recombinant vaccines over a longer period. No change is required in some countries/regions such as countries in South America that had retained trivalent influenza vaccines throughout this period.

While it is unlikely that changing from quadrivalent vaccines to trivalent vaccines will make influenza vaccines cheaper (they are already among the cheapest available vaccines), this situation offers several intriguing opportunities. The two most obvious consequences are that more influenza vaccine doses could be made because there is one less component to manufacture or alternatively, the same number of vaccine doses could be made available in a shorter period of time. The availability of more vaccine could benefit developing countries, allowing them to introduce tailored influenza vaccination programs. On the other hand, if the same amount of vaccine that is currently produced can be produced more quickly, the vaccine recommendations made by WHO in late September and February each year could be shifted forward by a number of weeks. Delaying the vaccine composition decision by a few weeks would increase the amount of data available for review and could potentially improve the match between the recommended vaccine components and the viruses that circulate in the following season, although this was not apparent from recent study⁸.

Another option worth considering would be to maintain a quadrivalent vaccine where the fourth component could be an additional strain of any of the three components e.g., two A(H3N2), or two A(H1N1)pdm09 or two B/Vic viruses, instead of the current single representative component of each virus. This approach has the potential to both enhance and broaden the vaccine-induced immune response and could be flexible, with a selection of the fourth component as needed. A similar outcome might also be achieved by simply increasing the amounts of a particular component(s). For example, the current content of HA in most standard inactivated influenza vaccines is 15 µg per type/subtype which could be increased to 30 µg HA per dose for one component with the removal of the B/Yam component and this change would not exceed the current regulatory maximum protein content of human influenza vaccines. Increasing the amount of influenza antigens in the vaccine has resulted in improved immunogenicity and vaccine efficacy in the Sanofi High Dose influenza vaccine for the elderly, where all components of the vaccine are present at 60 µg HA⁹. Whether an additional antigen is added or the amount of antigen in each dose is increased, the argument for improving A(H3N2) vaccines is the strongest because vaccine effectiveness against A(H3N2) viruses has been poor for many years¹⁰.

What are the potential challenges with this approach? The addition of multiple strains from the same type/subtype or significantly increasing the amount of one or more viral antigens, would require clinical trials and regulatory approval. However, clinical studies could focus on serological outcomes rather than vaccine efficacy, as was the case when the second B component was added to the trivalent vaccine over a decade ago.

The removal of the B/Yam component may also improve the performance of mRNA-based vaccines as initial clinical trials showed lower serologic responses against the two influenza B HA components compared to conventional influenza vaccines. The removal of B/Yam from the vaccine could allow mRNA manufacturers to increase the amount of the B/Vic component to improve immune responses. However, this may not be necessary as it has been reported that other measures taken since the first trials have already succeeded in improving the antibody responses to vaccines containing one influenza B component¹¹.

Apart from the implications for vaccines, what other matters should be considered with the elimination of B/Yam viruses? It will be important to minimise the possibility of the re-introduction of B/Yam viruses into the human population. Importantly, as there is no permanent animal reservoir for influenza B viruses, it is not possible for the B/Yam lineage viruses to re-emerge from an animal source. With the upcoming removal of B/Yam from the Astra Zeneca-manufactured LAIV, the remote risk of the attenuated cold-adapted B/Yam virus reassorting with a circulating B/Vic virus in a vaccine recipient to re-establish a ‘wild-type’ B/Yam virus that could spread, will be removed. This leaves laboratory-acquired infections as the next most plausible mode of re-introduction of B/Yam. While many laboratories will have stocks of B/Yam lineage viruses in their freezers, laboratory-acquired infections with influenza viruses are relatively rare because influenza viruses are spread mainly through aerosols, and these are usually minimised and contained in laboratory settings. The risk of infection is further mitigated in laboratory staff because many have been previously vaccinated against influenza (a common practice in many clinical/influenza laboratories) and B/Yam has been in influenza vaccines until very recently. In all except immunologically naïve hosts, re-exposure to B/Yam virus will recall immune memory from prior infection or vaccination. There is also a level of cross-protection afforded by future B/Vic infections or vaccinations against B/Yam viruses as has been shown previously^12,13,14. Thus, an individual lab-acquired infection is unlikely to result in a major outbreak leading to global re-introduction of B/Yam. When levels of immunity to B/Yam eventually wane sufficiently in the general population, they will also benefit from cross-lineage protection and even in a worst-case scenario involving the re-emergence of B/Yam viruses, the virus is likely to spread relatively slowly and manufacturers could then re-introduce B/Yam into their influenza vaccines. Although it is likely that the level of cross-protection offered by B/Vic infections/vaccines may wane in coming years, continued global surveillance for outbreaks of B/Yam infections will be of greater value in assessing the risk of re-introduction of these viruses than periodic immune assays in selected populations. Influenza B viruses unlike influenza A viruses, cannot cause pandemics and so are less of a global threat, even if an old lineage might re-appear or a new B-lineage might emerge.

In conclusion, the apparent extinction of B/Yam is a positive outcome of the COVID-19 pandemic, possibly leading to fewer influenza B infections and allowing an opportunity for more flexibility in the way we approach influenza vaccine manufacture. Furthermore, while there is only a low risk that B/Yam viruses could be re-introduced to the global population, the impact of such an event is likely to be relatively minor and manageable. Nevertheless, scientists and laboratories should minimise the use and exposure of staff to samples and isolates containing B/Yam viruses and prepare for removal of stored viruses from general access, to help mitigate any chance of re-emergence of B/Yam viruses. A reassessment or review by an international panel should be convened in about 10 years to re-examine the handling and storage of B/Yam virus stocks in laboratories. How the absence of B/Yam affects the epidemiology and evolution of human influenza viruses in the longer term remains to be seen, but as far as the basics are concerned, it is unlikely to herald the future disappearance of B/Vic viruses or one of the two influenza A viruses that currently infect mankind. Whether a divergent influenza B virus emerges in the future, repeating the 1970’s influenza B virus split, is unknown but should be the focus of continued careful surveillance.