Waning of SARS-CoV-2 vaccine-induced immunity: A systematic review and secondary data analysis

Abstract

BackgroundThe emergence of Omicron (B.1.1.529) variant of SARS-CoV-2 in late 2021 was followed by a marked increase of breakthrough infections. Estimates of vaccine effectiveness (VE) in the long term are key to assess potential resurgence of COVID-19 cases in the future.MethodsWe conducted a systematic review of manuscripts published until June 21, 2022 to identify studies reporting the level of protection provided by COVID-19 vaccines against SARS-CoV-2 infection and symptomatic disease at different time points since vaccine administration. An exponential model was used to perform a secondary data analysis of the retrieved data to estimate the progressive waning of VE associated with different vaccine products, numbers of received doses, and SARS-CoV-2 variants.FindingsOur results show that VE of BNT162b2, mRNA-1273, ChAdOx1 nCoV-19 vaccines against any laboratory confirmed infection with Delta might have been lower than 70% at 9 months from second dose administration. We found a marked immune escape associated with Omicron infection and symptomatic disease, both after the administration of two and three doses. The half-life of protection against symptomatic infection provided by two doses was estimated in the range of 178-456 days for Delta, and between 66 and 73 days for Omicron. Booster doses were found to restore the VE to levels comparable to those acquired soon after administration of the second dose; however, a fast decline of booster VE against Omicron was observed, with less than 20% VE against infection and less than 25% VE against symptomatic disease at 9 months from the booster administration.ConclusionsThis study provides a cohesive picture of the waning of vaccine protection; obtained estimates can inform the identification of appropriate targets and timing for future COVID-19 vaccination programs.

Tracking the progressive spread of the SARS-CoV-2 Omicron variant in Italy, December 2021 to January 2022

“…The selective advantage of Omicron over Delta may be explained by increased transmissibility, partial immune escape, or a combination of both [ 15 ]. Early evidence from statistical analysis of reinfections and breakthrough infections suggested a marked capability of escape from natural and vaccine-acquired immunity [ 11 , 16 - 19 ]. However, further efforts are needed to quantify the relative transmissibility of Omicron compared with Delta, and its inherent capacity to cause severe disease, which is key to evaluate potential changes in COVID-19 burden potentially led by this and the progressive waning of immunity.…”

Endemic means change as SARS-CoV-2 evolves

“…While the rise in frequency of a variant, as described by its selective advantage (equation (4)) and the long-term impact of the variant on the endemic equilibrium (equation (5)) are insensitive to the number of recovered compartments ( n ), the extent of oscillations following the initial spread of the variant are much stronger as n increases (Figure S2, panels A-C). Empirically, the distribution of waning times is close to exponential (CV = 1; (Menegale et al 2023), suggesting that intrinsic oscillations are likely to be damped (like Figure S2A, where CV = 1). Similar behaviour to Figure S2A is seen in a model with only two recovered classes ( n = 2), corresponding to high (R 1 ) and low (R 2 ) antibody levels, where an immune evasive variant (but not the resident virus) can infect the second class. By setting the waning rates to δ 1 = δ/x (from R 1 to R 2 ) and δ H = δ /(1 − x ) (from R 2 to S), the equilibrium fraction of recovered individuals in the second class ( x ) can be adjusted to allow for more immune evasive variants, while keeping the average time from first recovering to susceptibility at 1/ δ days for the resident virus.…”

“…Source: Waning rate depends on the exact sequence of vaccinations and infections. The half-life of protection against symptomatic infection with Omicron among studies summarized by Menegale et al (Menegale et al 2023) was 87 days without a booster and 111 days with a booster, yielding δ values ranging from 0.0071 to 0.0094 per day. Waning rates were similar for older and younger individuals (Menegale et al’s eFigure 14). Î of 2% [range of 0.5%-4%].…”

Estimating SARS‐CoV‐2 infections and associated changes in COVID‐19 severity and fatality

“…26,33 Accordingly, we consider variant-specific average durations of protection after two doses of vaccine and after a booster dose. 34 Shorter and longer durations of vaccine protection are explored in the sensitivity analyses (Table S4). Further details on the model and parameter values are provided in the Supporting Information (Table S2).…”

“…Waning of vaccine protection during the ancestral and Alpha phases is not considered; indeed, ancestral lineages were replaced by the Alpha variant a few weeks after the start of the vaccination campaign in Italy, and the waning of vaccine protection estimated in the literature during dominance of the Alpha variant is negligible 26,33 . Accordingly, we consider variant‐specific average durations of protection after two doses of vaccine and after a booster dose 34 . Shorter and longer durations of vaccine protection are explored in the sensitivity analyses (Table S4).…”