Reply to Yager and Doll Free

To the Editor—We thank the authors for their interest in our study comparing effectiveness of quadrivalent and trivalent inactivated influenza vaccines (IIV4 and IIV3, respectively) against the 2 influenza B lineages (Yamagata and Victoria) from 2013–2014 to 2016–2017 [1]. We found that increasing uptake of IIV4 was not associated with significantly increased protection against influenza B illness overall (VE of 53% [95% confidence interval {CI}, 45%–59%] for IIV4 vs 45% [95% CI, 34%–54%] for IIV3) despite higher effectiveness of IIV4 against the cross-lineage B viruses not included in IIV3 (VE of 56% for IIV4 vs 25% for IIV3) [1]. During the 2019–2020 season, all influenza vaccines distributed in the United States are quadrivalent except for 2 licensed only for use in adults aged ≥ 65 years [2]. Assuming low cross-lineage B virus protection, Reed et al [3] estimated additional public health benefit of quadrivalent vaccines from 1999–2000 to 2008–2009 to be modest reductions in influenza-associated outcomes. Because recent data from the US Flu VE Network [1] and Canada [4] over multiple seasons showed low to modest IIV3 cross-lineage protection against influenza B, an updated analysis of the public health impact and cost-benefit of quadrivalent vs trivalent influenza vaccines is possible.

Several points in the letter by Yager and Doll merit consideration. First, there may be differences in prime and boost sequence by each B lineage in children [1, 5]. Since most vaccine used worldwide is trivalent [6], considering the potential of each B lineage virus for inducing cross-reactive antibodies in persons with preexisting antibodies [7] is important for strain selection of B lineage in trivalent vaccine. Vaccine effectiveness studies are important to determine cross-protection during B lineage mismatch seasons. Further study is needed to determine whether priming with both B lineages induces broader neutralizing antibodies along with memory B cells against both B lineages [8]. The need for developing universal influenza vaccine is underscored by the divergent evolution of B lineages with poorly cross-reactive hemagglutinin variants and antigenically drifted neuraminidase [9]. With limited sample sizes in our age-stratified analyses over the 4 seasons, including one B lineage mismatch season (2015–2016), for children aged 6 months to 17 years, IIV3 VE was lower but the 95% CIs overlapped (VE of 58% [95% CI, 43%–69%] for IIV4 vs 38% [95% CI, 2%–61%] for IIV3). However, adults aged 18–49 years who received IIV3 were not protected against cross-lineage B viruses, unlike those receiving IIV4 [1].

Because both children and young adults are more susceptible to influenza B, cost-effectiveness studies need to consider different age-based vaccination strategies [10]. Future effectiveness estimates of both quadrivalent live attenuated and inactivated influenza vaccines are needed. Although not replacing the direct vaccination benefit in adults, immunizing children is important for individual, household, and community protection [11]. Our study was designed to estimate direct effectiveness alone. Last, more comparative effectiveness studies of trivalent and quadrivalent vaccines against influenza B over multiple seasons from different countries could possibly lead to future meta-analyses and cost-effectiveness studies with targeted age-based vaccination strategies.

Notes

Disclaimer. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention (CDC).

Potential conflicts of interest. M. G. reports grants from the CDC, CDC-Abt, and MedImmune/Astra Zeneca. B. F. reports no potential conflicts of interest. Both authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

References