How effective are current COVID-19 vaccines (including updated bivalent/variant-adapted boosters) at preventing symptomatic infection with XBB.1.16, EG.5, and other dominant 2025 variants?
Executive summary
Updated, variant-adapted COVID-19 vaccines (XBB.1.5-, JN.1- and KP.2-targeted monovalent boosters and prior BA.4/5 bivalents) reduce symptomatic infection and strongly reduce severe outcomes early after receipt, but protection against symptomatic infection is variable across variants and wanes over months. Real‑world meta‑analyses and cohort studies report symptomatic‑infection VE roughly in the 20–53% range versus Omicron XBB‑lineages shortly after boosting and substantially higher protection against hospitalization (pooled VE ~56% in older adults for XBB.1.5‑adapted vaccines) while studies consistently document rapid waning by 3–5 months [1] [2] [3].
1. What the data say today: modest prevention of symptomatic infection, stronger protection against severe disease
Multiple systematic reviews and large observational studies show variant‑adapted mRNA boosters provide additional protection against symptomatic infection with XBB‑related sublineages but not complete protection. A network meta‑analysis and systematic review found XBB.1.5‑adapted mRNA vaccines were effective and comparable for XBB‑related infections, but VE estimates against infection were modest (many studies in the pooled analyses produced single‑digit to mid‑50s percentages) while protection against hospitalization was higher [1] [4]. A meta‑analysis of monovalent XBB.1.5 vaccines reported VE in month one of about 53% against infection with sharper declines by month five [3]. Among adults ≥65, pooled VE against hospitalization after XBB1.5‑adapted boosters was ~56% (95% CI, 51–60) [2].
2. Variant match and timing matter: better match gives better early VE, but immune escape and waning erode benefits
Studies comparing periods of different variant predominance show vaccine effectiveness tracks antigenic match. XBB‑era data tended to show higher VE than JN.1‑era data, and KP.2‑era results rebounded when vaccines matched circulating strains better [5] [6]. Randomized immunogenicity and real‑world analyses both show that variant‑matched boosters elicit higher neutralizing titers against their target lineage and initially higher VE, but neutralizing antibodies and clinical protection fall over 2–5 months, with several reports noting clear waning by 2–3 months for protection against XBB lineages [7] [3].
3. What "effective" means here — endpoints and populations shift the picture
Effectiveness against symptomatic infection is a different question than preventing hospitalization or death. Bivalent and variant‑adapted boosters generally show larger effect sizes against severe outcomes than against symptomatic infection (examples: bivalent VE against hospitalization ~61.8% vs much lower monovalent estimates; pooled hospitalization VE for XBB1.5 boosters ~56% in older adults) [8] [2]. Observational VE varies by age, comorbidity, prior infection and vaccine platform; several studies caution against simple cross‑study comparisons because timing, circulating sublineages, and prior immunity differ [9] [4].
4. Strengths and limits of the evidence: mostly real‑world, observational and immunogenicity surrogates
Regulators and WHO explicitly request variant‑specific, age‑stratified VE data to guide antigen selection, reflecting that much evidence is observational and evolving [10]. Many cited estimates come from test‑negative or cohort designs and meta‑analyses of observational studies; randomized, pre‑licensure efficacy trials of XBB‑ or KP.2‑targeted vaccines were not performed before some approvals, so regulators rely on immunogenicity, bridging and real‑world data [11] [12]. Neutralizing antibody titers correlate strongly with protection from symptomatic infection and rise after variant‑matched boosters, but antibody correlates wane and T‑cell protection is less well captured by these measures [13] [14].
5. Competing interpretations and policy implications
Some syntheses argue variant‑adapted boosters restore substantial protection against severe disease and provide useful short‑term reductions in symptomatic infection; others emphasize limited, rapidly waning VE against infection and call for next‑generation vaccines or alternate strategies (broader/bivalent, mucosal, or different antigen designs) and close monitoring of KP.2/JN.1-era performance [1] [11] [15]. WHO TAG‑CO‑VAC and national advisory bodies continue to request more variant‑specific VE data to inform antigen composition decisions for 2025 deliberations [10].
6. Bottom line for individuals and health systems
Available evidence shows variant‑adapted boosters reduce your chance of symptomatic infection with XBB‑lineage viruses only partially and temporarily, but they substantially reduce risk of hospitalization and death, especially in older and high‑risk people — benefits that fall as months pass and as new immune‑evasive lineages emerge [3] [2] [4]. Public health authorities are monitoring VE by variant and urging updated data sharing from manufacturers and researchers to inform future vaccine strain choices [10].
Limitations and transparency: this analysis is based solely on the supplied set of peer‑reviewed studies, preprints, reviews and agency documents; direct head‑to‑head randomized efficacy trials of many 2024–2025 variant‑adapted vaccines were not done before authorization, and several sources note scarce data for specific lineage comparisons such as JN.1 [11] [5]. Available sources do not mention long‑term (>6 months) randomized efficacy of KP.2 or later monovalent vaccines.