Download:
-
The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) Omicron variant has infected over 90% of the global population at least once (1). The protection conferred by previous infections is gradually becoming a crucial factor in controlling the pandemic (2). Our research used systematic reviews and meta-analyses to estimate the degree and longevity of protection against reinfection by another Omicron sub-lineage, relative to uninfected individuals, under a similar vaccination status. Out of 14,105 publications, we selected 10 studies that had either a cohort, test-negative design, or case-control approach, and utilized their data for a statistical analysis. Our findings indicate that the immunity provided against reinfection tends to vary based on the previous variant encountered and the variant causing reinfection. Moreover, protection offered by Omicron sub-lineage infection against reinfection with another Omicron sub-lineage tends to decrease over time. The degree of protection from a prior infection increases with more closely related antigenic distance and higher humoral immunity levels.
We employed three-level meta-analytic models using the ‘metagen’ function of the ‘meta’ package (version 6.3) in R (version 4.3.1, R Foundation for Statistical Computing, Vienna, Austria) for consolidating protection data. We extracted multiple protection data points from a single study, incorporating all in the meta-analyses. Three-level meta-analyses permit the explicit modeling of nested data structures, such as when individual studies provide multiple estimates for varying subgroups or time points. These models yield more valid and reliable estimates than traditional fixed and random-effect models under such conditions (3). For research data specifying time from initial infection, we applied a meta-regression of the log odds to approximate the waning of protection over time, assessing at 1-month intervals. We performed meta-analysis and meta-regression only on groups comprising more than two articles with verifiable extracted data. Database searches covered PubMed, the World Health Organization (WHO) coronavirus disease 2019 (COVID-19) database, SSRN, MedRxiv, Embase, and the WanFang Database. We searched for cohort, test-negative design, and case-control studies published on or before October 24, 2023, using keywords related to reinfection, prior infection, and Omicron (
Supplementary Table S1 ). Included studies were those that considered the protective effect of prior Omicron infection in individuals against those who were infection-naïve and had comparable vaccination status. We evaluated the risk of bias using the ROBINS-I tool (Supplementary Table S4 ). Our study, which complies with PRISMA, was registered with PROSPERO (CRD42023466200).Supplementary Materials provide detailed methodology.We reviewed the titles and abstracts of 14,105 articles, of which 491 passed our screening to undergo a thorough full-text review (
Supplementary Figure S1 ). From this process, we identified 10 relevant studies providing 81 data sets, sourced from eight nations: Qatar, Canada, China, Denmark, Japan, the Republic of Korea, Singapore, and Portugal. These studies encompassed a combined sample size of 17,214,915. For our meta-analysis and meta-regression, we included 12 data sets from 2 studies in the BA.1 to BA.2 group, 10 data sets from 3 studies in the BA.1 to BA.4/5 group, 12 data sets from 3 studies in the BA.2 to BA.4/5 group, and 15 data sets from 4 studies in the BA.1/2 to BA.4/5 group (Supplementary Table S2 ).Compared to a non-infected cohort, individuals previously infected with the BA.1 variant showed 87.5% protection (47.9–97.0) against reinfection with the BA.2 variant (
Supplementary Figure S2 ). This protection, however, waned from 89.8% at 1 month (64.6–97.1) to 81.1% at 5 months (31.9–94.8) (Tables 1–2). Between 30 and 60 days post-infection, an 82.0% protection rate (49.0–94.0) was observed amongst those unvaccinated, and a protection rate of 94.2% (89.2–96.9) amongst those vaccinated (Supplementary Table S2 ). In vaccinated individuals, the effectiveness of protection against reinfection with BA.4/5 variant after an initial BA.1 infection was 75.2% (42.1–89.4) (Supplementary Figure S2 ). Notably, this protection waned from 77.2% at 5 months post-infection (47.6–90.1) to 40.9% at 12 months (−32.8–73.7) (Tables 1–2).Type of variant for prior infection Type of variant for reinfection Time since primary infection BA.2 (Meta-analysis) BA.4/5 (Meta-analysis) BA.2.75 (Systematic review) XBB (Systematic review) BA.1 5 months 81.1 (31.9–94.8) 77.2 (47.6–90.1) NA 40.0 (32.0–47.0)† 8 months NA 65.7 (26.5–84.0) NA 27.0 (24.0–30.0)§ Total* 87.5 (47.9–97.0) 75.2 (42.1–89.4) NA NA BA.2 5 months NA 89.6 (76.9–95.4) NA 74.0 (72.0 to 75.0)¶ 8 months NA 80.4 (56.7–91.1) NA 37.0 (32.0 to 43.0)** Total* NA 88.9 (76.6–94.8) NA NA BA.1/2 Total* NA 86.2 (73.6–92.8) 49.9 (47.6 to 52.1) NA BA.4/5 Total* NA NA 80.6 (71.2 to 87.0) NA Note: NA means no data available.
* Total protection, regardless of time since primary infection.
† Time since primary infection: 3–8 months.
§ Time since primary infection: ≥8 months.
¶ Time since primary infection: 3–6 months.
** Time since primary infection: ≥8 months.Table 1. Protection (%) against reinfection by different Omicron sub-lineages.
Time since primary infection (months) BA.1 to BA.2 BA.1 to BA.4/5 BA.2 to BA.4/5 1 89.8 (64.6–97.1) NA NA 2 88.1 (60.6–96.4) NA NA 3 86.1 (54.5–95.8) NA NA 4 83.8 (45.3–95.2) NA 91.6 (80.9–96.3) 5 81.1 (31.9–94.8) 77.2 (47.6–90.1) 89.6 (76.9–95.4) 6 NA 73.9 (41.9–88.3) 87.2 (71.8–94.2) 7 NA 70.1 (35.0–86.2) 84.1 (65.2–92.8) 8 NA 65.7 (26.5–84.0) 80.4 (56.7–91.1) 9 NA 60.1 (16.1–81.6) NA 10 NA 55.0 (3.2–79.1) NA 11 NA 48.5 (−12.8–76.4) NA 12 NA 40.9 (−32.8–73.7) NA Note: NA means no data available. Table 2. Estimates of protection (%) against various Omicron variants based on the time elapsed since primary infection.
Research conducted in Singapore (4) tracked the infection history of cohorts unexposed to COVID-19 who later contracted the BA.1 variant. Observations indicated that the protective effect of prior BA.1 infection against clinically attended symptomatic XBB variant reinfection diminished from a span of 40.0% (32.0–47.0) between 3–8 months post-infection to 27.0% (24.0–30.0) subsequent to 8 months (Table 1,
Supplementary Table S2 ).The protective effect of a BA.2 infection against subsequent reinfection with BA.4/5 variants was 88.9% (76.6–94.8) (
Supplementary Figure S2 ), and this waned from 91.6% (80.9–96.3) at 4 months post-infection to 80.4% (56.7–91.1) at 8 months post-infection (Tables 1–2). Comparable findings were reported by a Singaporean cohort study (4), which demonstrated that the protective effect of a primary BA.2 infection against symptomatic reinfection by the XBB variant also declined over time, from 74.0% (72.0–75.0) during the 3–6 months post-infection period to 37.0% (32.0–43.0) after 8 months (Table 1, Supplementary Table S2). It is important to note that these studies were conducted exclusively among vaccinated populations due to a lack of data from unvaccinated groups.Group Country Study Design Vaccination Status Outcome of reinfection Protection (95% CI) Time since primary infection (reported) Time since primary infection (predicted)* sample size BA.1 to BA.2 Chemaitelly et al. (2022) (1) Qatar TND Vaccinated (adjusted) Infection 94.2 (89.2 to 96.9) 39 to 45 days NA 41,988 Carazo et al.
(2023) (2)Canada Cohort Unvaccinated Infection 82.0 (49.0 to 94.0) 30 to 59 days NA 2,567 Carazo et al.
(2023) (2)Canada Cohort Unvaccinated Infection 76.0 (63.0 to 85.0) 60 to 89 days NA 2,567 Carazo et al.
(2023) (2)Canada Cohort Unvaccinated Infection 70.0 (61.0 to 77.0) 90 to 182 days NA 2,567 Carazo et al.
(2023) (2)Canada Cohort Unvaccinated Infection 72.0 (65.0 to 78.0)§ NA 81 2,567 Carazo et al.
(2023) (2)Canada Cohort Unvaccinated Symptomatic infection 86.0 (79.0 to 91.0)§ NA 81 630 Carazo et al.
(2023) (2)Canada Cohort Unvaccinated Infection 67.0 (79.0 to 91.0)§ NA 81 1,167 Carazo et al.
(2023) (2)Canada Cohort Unvaccinated Infection 76.0 (65.0 to 83.0)§ NA 81 1,258 Carazo et al.
(2023) (2)Canada Cohort Unvaccinated Infection 73.0 (65.0 to 79.0)§ NA 81 2,416 Carazo et al.
(2023) (2)Canada Cohort Unvaccinated Infection 72.0 (65.0 to 78.0)§ NA 81 2,567 Carazo et al.
(2023) (2)Canada Cohort Unvaccinated Infection 72.0 (63.0 to 79.0)§ NA 81 2,040 Carazo et al.
(2023) (2)Canada Cohort Unvaccinated Symptomatic infection 88.0 (80.0 to 93.0)§ NA 81 433 BA.1 to BA.4/5 Chen et al.
(2023) (3)China Cohort Vaccinated (matched) Symptomatic infection 64.1 (52.4 to 73.1) 329 to 341 days NA 386 Tan et al.
(2023) (4)Singapore Cohort Completed primary series Medically attended symptomatic infection 55.0 (17.0 to 76.0) NA 265 1,733,535 Tan et al.
(2023) (4)Singapore Cohort Boosted Medically attended symptomatic infection 61.0 (54.0 to 67.0) NA 265 1,902,581 Tan et al.
(2023) (4)Singapore Cohort Vaccinated (adjusted) Medically attended symptomatic infection 65.0 (57.0 to 72.0) 3 to <8 months NA 1,779,968 Tan et al.
(2023) (4)Singapore Cohort Vaccinated (adjusted) Medically attended symptomatic infection 49.0 (35.0 to 59.0) ≥8 months NA 1,858,209 Tan et al.
(2023) (4)Singapore Cohort Completed primary series Medically attended symptomatic infection 40.0 (-31.0 to 72.0)§ NA 265 1,351,636 Tan et al.
(2023) (4)Singapore Cohort Boosted Medically attended symptomatic infection 63.0 (55.0 to 69.0)§ NA 265 1,514,582 Tan et al.
(2023) (4)Singapore Cohort Vaccinated (adjusted) Medically attended symptomatic infection 66.0 (57.0 to 73.0)§ 3 to <8 months NA 1,402,809 Tan et al.
(2023) (4)Singapore Cohort Vaccinated (adjusted) Medically attended symptomatic infection 51.0 (37.0 to 62.0)§ ≥8 months NA 1,465,449 Jang et al.
(2023) (5)South Korea Case-control Vaccinated (adjusted) Infection 89.5 (89.2 to 89.8) NA 190 5,085,535 BA.1 to XBB Tan et al.
(2023) (4)Singapore Cohort Completed primary series Medically attended symptomatic infection 3.00 (−13.0 to 16.0) NA NA 800,221 Tan et al.
(2023) (4)Singapore Cohort Boosted Medically attended symptomatic infection 30.0 (27.0 to 32.0) NA NA 878,615 Tan et al.
(2023) (4)Singapore Cohort Vaccinated (adjusted) Medically attended symptomatic infection 40.0 (32.0 to 47.0) 3 to <8 months NA 796,534 Tan et al.
(2023) (4)Singapore Cohort Vaccinated (adjusted) Medically attended symptomatic infection 27.0 (24.0 to 30.0) ≥8 months NA 883,260 Tan et al.
(2023) (4)Singapore Cohort Completed primary series Medically attended symptomatic infection 16.0 (−6.0 to 33.0)§ NA NA 630,473 Tan et al.
(2023) (4)Singapore Cohort Boosted Medically attended symptomatic infection 30.0 (27.0 to 33.0)§ NA NA 706,028 Tan et al.
(2023) (4)Singapore Cohort Vaccinated (adjusted) Medically attended symptomatic infection 41.0 (32.0 to 49.0)§ 3 to <8 months NA 631,880 Tan et al.
(2023) (4)Singapore Cohort Vaccinated (adjusted) Medically attended symptomatic infection 28.0 (25.0 to 31.0)§ ≥8 months NA 705,569 BA.2 to BA.4/5 Chen et al.
(2023) (3)China Cohort Vaccinated (matched) Symptomatic infection 88.6 (81.7 to 93.1) 210 to 231 days NA 346 Tan et al.
(2023) (4)Singapore Cohort Completed primary series Medically attended symptomatic infection 80.0 (53.0 to 91.0) NA 181 1,737,378 Tan et al.
(2023) (4)Singapore Cohort Boosted Medically attended symptomatic infection 78.0 (74.0 to 82.0) NA 181 2,124,162 Tan et al.
(2023) (4)Singapore Cohort Vaccinated (adjusted) Medically attended symptomatic infection 87.0 (82.0 to 90.0) 3 to <6 months NA 1,866,720 Tan et al.
(2023) (4)Singapore Cohort Vaccinated (adjusted) Medically attended symptomatic infection 80.0 (74.0 to 84.0) 6 to <7 months NA 1,805,491 Tan et al.
(2023) (4)Singapore Cohort Vaccinated (adjusted) Medically attended symptomatic infection 74.0 (66.0 to 80.0) 7 to <8 months NA 1,885,447 Tan et al.
(2023) (4)Singapore Cohort Completed primary series Medically attended symptomatic infection 76.0 (24.0 to 93.0)§ NA 181 1,352,984 Tan et al.
(2023) (4)Singapore Cohort Boosted Medically attended symptomatic infection 79.0 (74.0 to 82.0)§ NA 181 1,743,385 Tan et al.
(2023) (4)Singapore Cohort Vaccinated (adjusted) Medically attended symptomatic infection 86.0 (82.0 to 90.0)§ 3 to <6 months NA 1,492,493 Tan et al.
(2023) (4)Singapore Cohort Vaccinated (adjusted) Medically attended symptomatic infection 80.0 (74.0 to 85.0)§ 6 to <7 months NA 1,430,362 Tan et al.
(2023) (4)Singapore Cohort Vaccinated (adjusted) Medically attended symptomatic infection 74.0 (66.0 to 80.0)§ 7 to <8 months NA 1,501,919 Jang et al.
(2023) (5)South Korea Case-control Vaccinated (adjusted) Infection 94.3 (94.1 to 94.4) NA 106 5,002,210 BA.2 to XBB Tan et al.
(2023) (4)Singapore Cohort Completed primary series Medically attended symptomatic infection 22.0 (9.0 to 33.0) NA NA 802,046 Tan et al.
(2023) (4)Singapore Cohort Boosted Medically attended symptomatic infection 51.0 (49.0 to 53.0) NA NA 982,831 Tan et al.
(2023) (4)Singapore Cohort Vaccinated (adjusted) Medically attended symptomatic infection 74.0 (72.0 to 75.0) 3 to <6 months NA 855,858 Tan et al.
(2023) (4)Singapore Cohort Vaccinated (adjusted) Medically attended symptomatic infection 58.0 (55.0 to 61.0) 6 to <7 months NA 820,235 Tan et al.
(2023) (4)Singapore Cohort Vaccinated (adjusted) Medically attended symptomatic infection 49.0 (47.0 to 52.0) 7 to <8 months NA 881,230 Tan et al.
(2023) (4)Singapore Cohort Vaccinated (adjusted) Medically attended symptomatic infection 37.0 (32.0 to 43.0) ≥8 months NA 809,865 Tan et al.
(2023) (4)Singapore Cohort Completed primary series Medically attended symptomatic infection 22.0 (−10.0 to 39.0) § NA NA 631,104 Tan et al.
(2023) (4)Singapore Cohort Boosted Medically attended symptomatic infection 51.0 (49.0 to 53.0) § NA NA 810,325 Tan et al.
(2023) (4)Singapore Cohort Vaccinated (adjusted) Medically attended symptomatic infection 74.0 (72.0 to 76.0) § 3 to <6 months NA 689,979 Tan et al.
(2023) (4)Singapore Cohort Vaccinated (adjusted) Medically attended symptomatic infection 59.0 (55.0 to 62.0) § 6 to <7 months NA 654,656 Tan et al.
(2023) (4)Singapore Cohort Vaccinated (adjusted) Medically attended symptomatic infection 50.0 (47.0 to 52.0) § 7 to <8 months NA 710,310 Tan et al.
(2023) (4)Singapore Cohort Vaccinated (adjusted) Medically attended symptomatic infection 37.0 (31.0 to 43.0) § ≥8 months NA 643,953 BA.1/2 to BA.4/5 Yamamoto et al.
(2023) (6)Japan Cohort All vaccinated (92% 3 doses) Infection 91.4 (73.2 to 97.3) NA NA 2,368 Altarawneh et al. (2022) (7)† Qatar TND Vaccinated (matched) Infection 80.5 (78.8 to82.0)§ 168 to 193 days NA 65,853 Altarawneh et al. (2022) (7)† Qatar TND Unvaccinated Infection 68.7 (64.0 to 72.9)§ 167 to 190 days NA 22,850 Altarawneh et al. (2022) (7)† Qatar TND All vaccinated Infection 83.7 (82.0 to 85.2)§ 170 to 194 days NA 43,003 Altarawneh et al. (2022) (7)† Qatar TND Vaccinated (adjusted) Symptomatic infection 76.3 (66.6 to 83.2)§ 175 to 196 days NA 2,838 Altarawneh et al. (2022) (7)† Qatar TND Vaccinated (adjusted) Infection 78.1 (75.1 to 80.7)§ 169 to 193 days NA 23,125 Altarawneh et al. (2022) (7)† Qatar TND Vaccinated (adjusted) Symptomatic infection 84.5 (81.1 to 87.2)§ 168 to 195 days NA 12,363 Altarawneh et al. (2022) (7)† Qatar TND Vaccinated (adjusted) Infection 80.3 (78.8 to 81.7)§ 168 to 193 days NA 77,399 Malato et al. (2022) (8) Portugal Cohort All vaccinated (98% 2 doses) Infection 75.3 (75.0 to 75.6) NA NA 6,885,922 Malato et al. (2022) (8) Portugal Cohort All vaccinated (98% 2 doses) Infection 76.8 (76.5 to 77.1)§ NA NA 6,279,978 Hansen et al.
(2023) (9)Denmark Case-control Three mRNA doses Infection 92.7 (91.6 to 93.7) NA NA 187,347 Hansen et al.
(2023) (9)Denmark Case-control Three mRNA doses Infection 92.9 (91.6 to 93.7)§ NA NA 17,238 Hansen et al.
(2023) (9)Denmark Case-control Three mRNA doses Infection 92.5 (91.4 to 93.5)§ NA NA 104,339 Hansen et al.
(2023) (9)Denmark Case-control Three mRNA doses Infection 94.4 (93.8 to 95.0)§ NA NA 219,643 Hansen et al.
(2023) (9)Denmark Case-control Three mRNA doses Infection 92.8 (91.7 to 93.7)§ NA NA 187,347 BA.1/2 to BA.2.75 Chemaitelly et al.
(2023) (10)Qatar TND Vaccinated (matched) Infection 49.9 (47.6 to 52.1) NA NA 105,431 Chemaitelly et al.
(2023) (10)Qatar TND Vaccinated (matched) Symptomatic infection 50.2 (43.1 to 56.4) NA NA 13,099 Chemaitelly et al.
(2023) (10)Qatar TND Vaccinated (matched) Infection 47.4 (44.8 to 49.8)§ NA NA 105,431 Chemaitelly et al.
(2023) (10)Qatar TND Vaccinated (matched) Symptomatic infection 49.7 (42.3 to 56.1)§ NA NA 13,099 Chemaitelly et al.
(2023) (10)Qatar TND Unvaccinated Infection 32.2 (25.5 to 38.3)§ NA NA 35,577 Chemaitelly et al.
(2023) (10)Qatar TND All vaccinated Infection 53.9 (51.5 to 56.3)§ NA NA 69,854 BA.4/5 to BA.2.75 Chemaitelly et al.
(2023) (10)Qatar TND Vaccinated (matched) Infection 80.6 (71.2 to 87.0) NA NA 102,271 Chemaitelly et al.
(2023) (10)Qatar TND Vaccinated (matched) Symptomatic infection 91.4 (35.8 to 98.8) NA NA 12,680 Chemaitelly et al.
(2023) (10)Qatar TND Vaccinated (matched) Infection 79.4 (69.4 to 86.2)§ NA NA 102,271 Chemaitelly et al.
(2023) (10)Qatar TND Vaccinated (matched) Symptomatic infection 91.3 (35.0 to 98.8)§ NA NA 12,680 Chemaitelly et al.
(2023) (10)Qatar TND Unvaccinated Infection 44.4 (−4.0 to 70.3)§ NA NA 34,862 Chemaitelly et al.
(2023) (10)Qatar TND All vaccinated Infection 87.4 (78.7 to 92.5)§ NA NA 67,409 Note: “NA” means not applicable.
Abbreviation: TND=Test-negative design.
* Time since primary infection was determined based on GISAID data, with 50% serving as the judgment standard for epidemic strain. A variant was considered dominant if it exceeds 50%.
† Based on the research start date and the definition of interval of reinfection, if the most recent infection occurred during the BA.1/2 dominant period, the variant for prior infection was considered to belong to BA.1/2.
§ Sensitivity analysis results of the original literature.Table S2. Characteristics of the included studies.
An analysis of protection against reinfection with the Omicron BA.4/5 variants following a BA.1/2 infection incorporated studies from four nations: Denmark, Japan, Portugal, and Qatar. Due to a gradual shift from the dominance of the Omicron BA.1 subvariant to the BA.2 subvariant in these countries, the two infection peaks of BA.1 and BA.2 combined, making it difficult to distinguish between their timelines. Our meta-analysis findings indicated that the conferred protection against reinfection with BA.4/5 after a BA.1/2 infection was 86.2% (73.6–92.8) in comparison to an uninfected population (
Supplementary Figure S2 ). Notably, a test-negative design study conducted in Qatar (5) reported protection rates of 49.9% (47.6–52.1) against a BA.2.75 infection after a primary BA.1/2 infection; this protection rate decreased to 32.2% (25.5–38.3) for an unvaccinated population. The study also reported protection rates of 50.2% (43.1–56.4) against symptomatic infection. Furthermore, a primary BA.4/5 infection offered a protection rate of 80.6% (71.2–87.0) against a BA.2.75 variant infection; this dropped to 44.4% (−4.0–70.3), however, for the unvaccinated population. The reported protection against symptomatic infection was 91.4% (35.8–98.8) (Table 1,Supplementary Table S2 ).According to meta-regression analyses of studies noting the duration since the initial infection, we discerned a decline in immunity against reinfection over time. However, due to limitations in available data, these estimated protection rates yielded wide confidence intervals (CIs), preventing any statistically significant differences from being determined within the meta-regression results. In spite of overlapping CIs within the meta-regression, at the same time since primary infection, BA.2 variant showed a higher protection estimate than BA.5 against BA.1 variant infection [at 5 months: 89.6% (76.9, 95.4) versus 77.2% (47.6, 90.1); at 8 months: 80.4% (56.7, 91.1) versus 65.7% (26.5, 84.0)]. A similar trend was observed when comparing immunity from BA.1 against BA.2, and BA.1 against BA.4/5 [at 5 months: 81.1% (31.9–94.8) versus 77.2% (47.6–90.1)] (Table 1-2, Figure 1,
Supplementary Table S3 ).
Figure 1.Estimated levels of protection against various Omicron variants based on the time elapsed since primary infection.
Note: The shaded areas represent the 95% confidence intervals (CIs). The size of each bubble is proportional to the reciprocal of the standard error (SE). -
The likelihood of reinfection with the same Omicron variant is exceedingly low in instances where a previous infection has occurred. In two respective studies, there was an impressive 94%–97% immunity rate observed amongst individuals previously infected with the BA.4/5 strain, which remained effective for up to 100 days post primary infection (6–7). Our comprehensive review and meta-analysis revealed elevated protection rates against reinfection in individuals previously exposed to later, evolutionarily similar variants, as compared to those exposed to earlier strains. This was independent of the time elapsed since the initial infection. This conclusion is congruent with studies conducted in the Netherlands (8), Qatar (5), Singapore (4), and the United Kingdom (9). This could potentially elucidate the cyclical nature of the COVID-19 pandemic, characterized by repeated bouts of infections and reinfections, triggered by distinct variants in succession.
The variants BA.4 and BA.5 are derivatives of BA.2. In terms of antigenic distance, BA.4/5 shares a closer resemblance with BA.2 relative to BA.1. Serological research has revealed that after infection by the BA.2 variant, the convalescent sera possess a superior amount of neutralizing antibodies against BA.5 than the sera derived from BA.1 variant infection (10). This collective evidence suggests that the effectiveness of protection from prior COVID-19 infection against further infections is not solely dependent on declining immunity but is also influenced by viral evolution. Higher humoral immunity levels and closer antigenic distances contribute to enhanced protection provided by previous infections.
In assessing the future risks associated with COVID-19, it is crucial to consider not only the time elapsed since the peak of the previous wave but also the antigenic difference and evasion ability of humoral immunity between any potential new variant and the formerly dominant ones. Ensuring a timely update of the antigenic components within the COVID-19 vaccine, coupled with inoculating individuals not recently infected, stands as a vital strategy in combating this disease. While this approach may not fully synchronize with the evolution of SARS-CoV-2, potentially leading to a mismatch between subsequent infection strains and the vaccine strain, a narrower antigenic distance can ostensibly offer improved protection over a match with a more antigenically distant strain. Given that both infection- and vaccine-induced protection diminish over time, the duration since either the infection or vaccination must be factored into vaccination policy considerations.
This study was subject to at least two limitations. One limitation of our research is that the time since primary infection provided by some studies is a time range. In order to conduct meta-regression, we used the median of this range. Another limitation lies in the limited number of studies incorporated. Upon retrieval and examination, only ten studies pertaining to the protection against Omicron variant infection satisfied the inclusion criteria. To offset potential overemphasis arising from the inclusion of numerous data points from a single study, we employed a three-tier meta-analysis approach. Nonetheless, as the available data pool was relatively insufficient, stratified analyses could not be conducted. Some specialists assert that a minimum of ten studies is required to facilitate valid meta-regression, which is greater than what we currently have. The scarcity of data broadens CIs, hampers the extraction of useful statistical inferences, and adds uncertainty to the stability of our final findings.
The global population has previously encountered pandemics involving the BA.1, BA.2, and BA.4/5 variants, with the infectious strain now transitioning to the XBB sub-lineage. While these past variant pandemics have subsided, further exploration of existing literature can deepen our understanding of the immune mechanisms underlying COVID-19. This will assist in securing epidemiological parameters of the disease, comprehending the mechanisms behind COVID-19 outbreaks, and providing essential evidence for infectious disease model research and assessments of reinfection risk.
Our comprehension of COVID-19 continues to evolve, highlighting the need for critical research into topics such as antigenic separation, immune evasion, and the extent of cross-protection. While predicting the trajectory of SARS-CoV-2 mutation remains a formidable challenge, establishing a link between pathogenesis and immunology through empirical research might expedite and enhance the precision of risk assessments for new variants. Understanding the degree of protection provided by previous COVID-19 infections against new variants can further inform and guide national response strategies.
-
No conflicts of interest.
-
Kathy Leung from the University of Hong Kong for her insightful suggestions on epidemiological concepts and statistical methods. Lance Rodewald from China CDC for his assistance with English language editing.
HTML
| Citation: |
|