[Skip to main content](/content/review/5ac7d462-1a70-4029-9382-e6d9f50ce22d/source/ss-268260010#main-content/index.html) Impact of Variants on Elasomeran Mechanism Research reportView only Create alertChat May 5, 2026 # How do updated formulations (variants of concern) affect the mechanism of Elasomeran? ## Updated mRNA-1273 formulations shift the neutralizing antibody response toward enhanced variant-specific recognition while maintaining the core B cell memory and T cell activation mechanisms of the original vaccine platform, with bivalent versions balancing variant-targeted depth against cross-reactive breadth. # Abstract Updated mRNA-1273 formulations targeting SARS-CoV-2 variants of concern alter the immune response mechanism through differential neutralizing antibody profiles while maintaining the platform’s core immunological pathways. Monovalent variant-specific vaccines (mRNA-1273.351 targeting B.1.351, mRNA-1273.529 targeting BA.1) demonstrated superior neutralizing antibody responses against their target variants, with geometric mean ratios of 1.67-1.77 for BA.1 formulations and twofold improvements for B.1.351 formulations, but showed reduced cross-reactivity against ancestral strains with fourfold lower neutralization against WA1/2020 D614G. Bivalent formulations (mRNA-1273.211, mRNA-1273.214) balanced this trade-off by maintaining or improving ancestral strain neutralization (GMR 1.11 against D614G for BA.1-bivalent) while achieving GMT ratios ≥1 against all variants of concern tested. All updated formulations induced robust anamnestic B cell memory responses when used as boosters, with variant-modified boosters showing 34.9-61.6 fold increases in neutralization against target variants, restoring immunity that had waned 6-7 fold against wild-type and 24-69 fold against variants over 6 months. Cellular immune mechanisms, including CD4+ and CD8+ T cell responses measured in preclinical models, showed similar activation patterns to the original vaccine, with effector T cell expansion but not effector-memory phenotypes and waning within 6 months. However, longitudinal studies revealed progressive viral escape from updated vaccine-induced immunity, with newer variants continuing to evade even XBB.1.5-updated formulations, indicating that antigenic drift outpaces the vaccine update cycle despite mechanistic improvements in neutralizing breadth and depth. Methods We analyzed 10 sources from an initial pool of 200, using 8 screening criteria. Each paper was reviewed for 8 key aspects that mattered most to the research question. More on methods Records from Elicit search n = 200 Papers screened using: Vaccine Type and Formulation, Mechanistic Outcomes, Comparative or Descriptive Data, Study Design, Updated Formulation Focus, Elasomeran Relevance, Mechanistic Data Inclusion, Study Rigor n = 200 Papers screened out n = 190 Papers included for extraction n = 10 Press enter or space to select a node.You can then use the arrow keys to move the node around. Press delete to remove it and escape to cancel. Press enter or space to select an edge. You can then press delete to remove it or escape to cancel. ## Paper search We performed a semantic search across over 138 million academic papers from the Elicit search engine, which includes all of [Semantic Scholar](https://www.semanticscholar.org/) and [OpenAlex](https://openalex.org/). We ran this query: “How do updated formulations (variants of concern) affect the mechanism of Elasomeran?” The search returned 200 total results from Elicit. We retrieved 200 papers most relevant to the query for screening. ## Screening We screened in sources based on their abstracts that met these criteria: - **Vaccine Type and Formulation**: Does this study investigate Elasomeran (mRNA-1273, Moderna COVID-19 vaccine) updated formulations specifically designed to target SARS-CoV-2 variants of concern (Alpha, Beta, Gamma, Delta, Omicron, or subsequent WHO-designated variants)? - **Mechanistic Outcomes**: Does this study report mechanistic outcomes including immunogenicity, antibody responses, T-cell responses, vaccine efficacy, or molecular mechanisms of action? - **Comparative or Descriptive Data**: Does this study provide comparative data between updated formulations and original formulations, other updated formulations, or descriptive mechanistic data on updated formulations? - **Study Design**: Is this study a randomized controlled trial, cohort study, case-control study, cross-sectional study, in vitro study, animal study, systematic review, or meta-analysis? - **Updated Formulation Focus**: Does this study investigate updated variant formulations (rather than focusing solely on the original Elasomeran formulation without investigation of updated variants)? - **Elasomeran Relevance**: Does this study investigate Elasomeran or include comparison to Elasomeran (rather than focusing solely on other COVID-19 vaccines like Pfizer-BioNTech, Johnson & Johnson, or AstraZeneca without Elasomeran investigation)? - **Mechanistic Data Inclusion**: Does this study include mechanistic investigation (rather than reporting only safety, adverse events, epidemiological outcomes, or logistical data without mechanistic components)? - **Study Rigor**: Is this study a full research article with adequate detail (rather than a case report, case series, editorial, commentary, or conference abstract)? We considered all screening questions together and made a holistic judgement about whether to screen in each paper. ## Data extraction We asked a large language model to extract each data column below from each paper. We gave the model the extraction instructions shown below for each column. - **Updated Formulation**: Extract details about the updated mRNA vaccine formulation studied that targets variants of concern, including: - Specific vaccine name/code (e.g., mRNA-1273.351, mRNA-1273.211, mRNA-1273.529) - Variant(s) targeted by the updated formulation (e.g., B.1.351/Beta, BA.1, XBB.1.5) - Type of formulation (monovalent variant-specific, bivalent original+variant, multivalent) - Dose amount (e.g., 50 μg, 25 μg) - How it differs from original mRNA-1273 - **Immunological Mechanisms**: Extract all immunological mechanisms measured to assess how the updated formulation works, including: - Neutralizing antibody responses (method of measurement, pseudovirus vs live virus) - Binding antibody responses (ELISA, immunoassays) - T-cell responses (CD4+, CD8+, cytokine production) - Cellular immune responses (memory B cells, plasma cells) - Durability measures (persistence over time) - Any mechanistic pathway analysis (complement activation, Fc function, etc.) - **Variant Testing Panel**: Extract which SARS-CoV-2 variants the immune mechanisms were tested against, including: - Ancestral/original strain (Wuhan-Hu-1, D614G, WA1/2020) - Specific variants of concern tested (Alpha, Beta, Gamma, Delta, Omicron sublineages) - Nomenclature used (WHO names, Pango lineages, spike mutations) - Whether testing was done against the variant the updated formulation was designed for vs other variants (cross-reactivity) - **Comparative Mechanism Results**: Extract quantitative comparisons of immune mechanisms between original mRNA-1273 and updated formulations, including: - Geometric mean titers/ratios with confidence intervals - Fold-change differences in antibody levels - Statistical significance of differences (p-values, confidence intervals) - Superiority, non-inferiority, or inferiority results - Cross-variant protection differences - Breadth vs depth of immune responses - Direction of effect (better, worse, equivalent) for each mechanism - **Vaccination Regimen**: Extract vaccination context that could affect immune mechanisms, including: - Primary vaccination series vs booster dose administration - Number of prior doses before updated formulation - Timing between doses (interval from last dose) - Previous vaccine types received (homologous vs heterologous) - Concurrent administration with other vaccines - Schedule used (2-dose primary, 3rd dose booster, 4th dose, etc.) - **Breakthrough Protection**: Extract data on breakthrough infections and protection specifically related to updated formulations vs original, including: - Breakthrough infection rates by variant - Disease severity in breakthrough cases - Viral load differences during breakthrough - Protection against symptomatic vs asymptomatic infection - Effectiveness estimates comparing updated vs original formulations - Duration of protection differences - **Study Population**: Extract population characteristics that could affect immune mechanisms with updated formulations, including: - Age groups studied (pediatric, adult, elderly) - Prior SARS-CoV-2 infection status (naive vs previously infected) - Immunocompromised status - Time since last vaccination or infection - Geographic location (relevant for circulating variants) - Sample size for mechanism analyses - **Mechanism Durability**: Extract data on how long the immune mechanisms persist with updated formulations compared to original, including: - Time points measured (weeks, months post-vaccination) - Rate of antibody waning over time - Persistence differences between updated vs original formulations - Memory response durability - When peak responses occurred - Whether mechanism differences were maintained over time # Results ## Characteristics of Included Studies The systematic review identified 10 sources examining updated mRNA-1273 (Elasomeran) formulations targeting SARS-CoV-2 variants of concern. Full texts were available for 6 studies, while 4 were available as abstracts only. Study Full text retrieved? Updated formulation(s) Study type Population Primary focus Ivan T Lee et al., 2023 No BA.1-monovalent (mRNA-1273.529), BA.1-bivalent (mRNA-1273.214) Phase 3 RCT Adults ≥16 years, UK 3-month safety and immunogenicity of BA.1-containing boosters Kai Wu et al., 2021 No mRNA-1273.351, mRNA-1273.211 Preclinical (mice) Mice Variant vaccine neutralization as primary or booster series A. Choi et al., 2021 Yes mRNA-1273.351, mRNA-1273.211 Phase 2a clinical trial Healthy adults (mean ages 47.5-63.8 years), US Safety and immunogenicity of variant-modified boosters K. Wu et al., 2021 Yes mRNA-1273.351, mRNA-1273.211 Phase 2 clinical trial Adults ≥18 years, US Preliminary safety and immunogenicity of variant vaccine booster M. Koch et al., 2021 No mRNA-1273.351, mRNA-1273.211 Preclinical (mice) Mice Variant vaccine neutralization and cross-protection Baoling Ying et al., 2021 Yes mRNA-1273.351, mRNA-1273.211 Preclinical (mice) 129S2 and K18-hACE2 mice Protective activity against ancestral and variant strains Baoling Ying et al., 2021a Yes mRNA-1273.351 Preclinical (mice) 129S2 and K18-hACE2 mice Protection against variants with suboptimal immunity M. Echaide et al., 2023 Yes Bivalent vaccines (Comirnaty, Spikevax) targeting BA.4-5 Omicron Review Multiple populations including elderly, pediatric, immunocompromised Advantages and limitations of mRNA vaccines against variants A. Roederer et al., 2024 Yes XBB.1.5 monovalent booster Prospective cohort Adults (median 33) and elderly (median 72), COVID-naive Evolution of viral escape from vaccine immunity A. Pegu et al., 2021 No None (durability study of original mRNA-1273) Clinical immunogenicity study Vaccine recipients Durability of antibodies against variants toof Pageof The studies examined three main categories of updated formulations: monovalent variant-specific vaccines (mRNA-1273.351 targeting B.1.351, mRNA-1273.529 targeting BA.1, XBB.1.5 booster), bivalent original+variant vaccines (mRNA-1273.211 combining original and B.1.351, mRNA-1273.214 combining original and BA.1), and later-generation bivalent vaccines targeting Omicron BA.4-5 subvariants. The dose for clinical formulations was consistently 50 μg, while preclinical studies used high doses (5 μg) and low doses (0.25 μg) to model varying immune responses. ## Immunological Mechanisms Assessed Studies evaluated multiple immunological mechanisms to characterize how updated formulations affect the immune response to SARS-CoV-2 variants. Mechanism type Measurement methods Studies assessing Neutralizing antibodies Pseudovirus neutralization assays (lentivirus-based, VSV-based); live virus focus-reduction neutralization test (FRNT) All clinical and preclinical studies Binding antibodies ELISA measuring IgG against recombinant spike proteins Preclinical studies CD8+ T cell responses H-2b-restricted immunodominant peptides, IFN-γ production Preclinical studies CD4+ T cell responses H-2b-restricted immunodominant peptides, IFN-γ production; S-specific CD4 T cells Preclinical studies and review B cell memory Anamnestic response to boosting Clinical booster studies Durability Persistence at Day 29 and Month 3; up to 6 months post-vaccination Clinical studies toof Pageof Neutralizing antibody responses were the primary mechanism assessed across all studies. Clinical trials used pseudovirus neutralization assays, while preclinical studies employed live virus FRNT methods. The clinical studies by A. Choi et al. validated their neutralization assays against both lentivirus-based and VSV-based pseudovirus systems. Cellular immune responses were characterized primarily in preclinical models, with CD8+ and CD4+ T cell responses measured through peptide stimulation and interferon-γ production. Human studies noted the presence of S-specific CD4 and CD8 T cells, with expansion of effector T cells but not effector-memory phenotypes. B cell memory was inferred from anamnestic responses to booster doses in clinical trials. ## Variant Testing Panels Updated formulations were tested against comprehensive panels of SARS-CoV-2 variants to assess both homologous and cross-reactive immune responses. Study Ancestral strain Variants of concern tested Variant formulation designed for Cross-reactivity assessed Ivan T Lee et al., 2023 D614G BA.1 Omicron BA.1 Yes (D614G) Kai Wu et al., 2021 Wild-type B.1.351 (Beta) B.1.351 Yes (broad cross-variant) A. Choi et al., 2021 D614G B.1.351 (Beta), P.1 (Gamma), B.1.617.2 (Delta) B.1.351 Yes (multiple VOCs) K. Wu et al., 2021 Wild-type B.1.351, P.1 B.1.351 Yes (P.1) M. Koch et al., 2021 Wild-type B.1.351 (Beta) B.1.351 Yes (broad) Baoling Ying et al., 2021 WA1/2020 D614G B.1.1.7 (Alpha), B.1.351 (Beta), B.1.617.2 (Delta) B.1.351 Yes (multiple) Baoling Ying et al., 2021a WA1/2020 D614G, D614G/N501Y B.1.1.7 (Alpha), B.1.351 (Beta), B.1.617.2 (Delta) B.1.351 Yes (multiple) M. Echaide et al., 2023 D614G B.1.1.7 (Alpha), B.1.351 (Beta), B.1.617.2 (Delta), B.1.1.529 (Omicron) BA.4-5 Omicron Yes (multiple) A. Roederer et al., 2024 Wuhan-Hu-1, D614G, WA1/2020 Beta (K417N/E484K/N501Y), Gamma, Delta (L452R/P681R), Omicron (BA.5, BQ.1.1, JN.1) XBB.1.5 Yes (comprehensive panel) A. Pegu et al., 2021 Not mentioned B.1.1.7 (Alpha), B.1.351 (Beta), P.1 (Gamma), B.1.429 (Epsilon), B.1.526 (Iota), B.1.617.2 (Delta) None (original vaccine) Yes (multiple VOCs) toof Pageof All studies tested cross-reactive immunity beyond the homologous variant targeted by each updated formulation. The most challenging variant across studies was B.1.351 (Beta), which showed the greatest resistance to neutralization by both original and updated formulations. Studies testing B.1.617.2 (Delta) consistently found lower neutralizing titers against this variant compared to ancestral strains. The comprehensive panel tested by A. Roederer et al. included 131 individual mutations across 48 named variants, revealing progressive viral escape from vaccine-induced immunity over time. ## Effects of Updated Formulations on Neutralizing Antibody Responses ### Superiority Against Homologous Variants Updated formulations demonstrated superior neutralizing antibody responses against their target variants compared to the original mRNA-1273 vaccine. Study Updated formulation Target variant Comparison to original mRNA-1273 GMR or fold-change Statistical significance Ivan T Lee et al., 2023 BA.1-monovalent BA.1 Omicron Superior at Month 3 GMR 1.77 (96% CI 1.55-2.02) Yes Ivan T Lee et al., 2023 BA.1-bivalent BA.1 Omicron Superior at Month 3 GMR 1.67 (96% CI 1.54-1.81) Yes A. Choi et al., 2021 mRNA-1273.351 (50 μg) B.1.351 Numerically greater titers 34.9-fold increase Not confirmed A. Choi et al., 2021 mRNA-1273.211 (50 μg) B.1.351 Numerically greater titers 61.6-fold increase P<0.0001 K. Wu et al., 2021 mRNA-1273.351 B.1.351 Superior neutralization 1400 vs 864 ID50 GMT 35-fold increase Baoling Ying et al., 2021 mRNA-1273.351 B.1.351 Twofold higher GMTs Twofold vs mRNA-1273 Superiority noted Baoling Ying et al., 2021a mRNA-1273.351 B.1.351 Twofold higher GMTs Twofold vs mRNA-1273 Superior against B.1.351 toof Pageof The BA.1-containing boosters showed 67-77% higher neutralizing antibody titers against Omicron BA.1 compared to the original vaccine at 3 months post-booster. For B.1.351-targeted formulations, the variant-specific mRNA-1273.351 vaccine induced 35-fold increases in neutralization titers post-boost, with geometric mean titers approximately twofold higher than the original vaccine. The bivalent mRNA-1273.211 formulation showed particularly strong responses, with a 61.6-fold increase in neutralization against B.1.351. ### Cross-Variant Protection Updated formulations maintained varying levels of cross-protection against non-target variants. Formulation Ancestral strain response Non-target variant responses Direction of effect BA.1-monovalent GMR 0.80 (95% CI 0.71-0.90) vs D614G Non-inferior against D614G Maintained BA.1-bivalent GMR 1.11 (96% CI 1.03-1.18) vs D614G Non-inferior against D614G Slightly improved mRNA-1273.351 Fourfold lower GMTs vs WA1/2020 D614G Lower against historical strains Trade-off mRNA-1273 Better against WA1/2020 Less reduction vs B.1.351 Broader baseline mRNA-1273.211 46.4-fold increase vs D614G GMT ratio ≥1 against all VOCs/VOIs Broadest protection toof Pageof The bivalent formulations demonstrated the most balanced cross-variant protection. The BA.1-bivalent vaccine showed slight improvement over the original vaccine against the ancestral D614G strain (GMR 1.11), while the BA.1-monovalent showed a modest 20% reduction (GMR 0.80), both meeting non-inferiority criteria. The B.1.351-targeted monovalent vaccine (mRNA-1273.351) showed a trade-off, with fourfold lower neutralization against ancestral strains but twofold higher neutralization against B.1.351. In contrast, the bivalent mRNA-1273.211 formulation achieved broad cross-variant neutralization with GMT ratios ≥1 against all variants of concern tested and showed a 46.4-fold increase against the wild-type D614G virus. ### Booster Effects on Waning Immunity Booster doses with updated formulations restored and enhanced neutralizing antibody titers that had declined over time. Study Pre-booster titers Post-booster titers Magnitude of increase Timing A. Choi et al., 2021 Waned significantly vs peak at Month 1 mRNA-1273: 16.7-fold vs D614G Significant (P<0.0001) ~6 months post-primary A. Choi et al., 2021 Low/undetectable vs B.1.351, P.1, B.1.617.2 mRNA-1273.211: 61.6-fold vs B.1.351 Significant (P<0.0001) ~6 months post-primary K. Wu et al., 2021 Low/below quantification 6-8 months post-primary Similar to or higher than peak post-primary at 2 weeks post-boost Robust memory response 2 weeks post-boost toof Pageof Prior to booster administration, neutralizing antibodies against variants had waned substantially from peak levels observed one month after the primary series. Neutralization titers against B.1.351, P.1, and B.1.617.2 were either low or undetectable at approximately 6 months post-primary vaccination. Both original and variant-modified boosters induced anamnestic responses indicative of robust B cell memory, with post-booster titers reaching or exceeding peak titers measured after the primary series. Two weeks after booster vaccination, neutralization titers increased to levels similar to or higher than peak titers after primary vaccination for both wild-type and variant viruses. ## Effects on Cellular Immune Responses Preclinical studies characterized T cell responses to updated formulations, while clinical data on cellular immunity were limited. Mechanism Findings with updated formulations Comparison to original CD8+ T cells Responses measured via H-2b peptides and IFN-γ production Not directly compared CD4+ T cells Responses measured via H-2b peptides and IFN-γ production Not directly compared Effector T cells Expansion observed with mRNA vaccines Similar to original Effector-memory T cells Not expanded by mRNA vaccines Limitation of platform T cell durability Generally lost 6 months post-vaccination Similar to original IL-17 production Elevated concentrations indicating inflammatory response Shared mechanism toof Pageof Both original and updated mRNA formulations induced CD4+ and CD8+ T cell responses that were detectable in preclinical models. However, clinical observations indicated that mRNA vaccines activate S-specific CD4 and CD8 T cells but do not expand effector-memory phenotypes, a limitation that appears consistent across formulations. T cell responses generally waned within 6 months post-vaccination, though T cells with stem cell memory phenotypes could potentially persist longer. The inflammatory signature remained similar between original and updated formulations, with elevated IL-17 concentrations observed across mRNA vaccines. ## Durability of Immune Mechanisms The persistence of immune responses varied by formulation type and dose level. Study Formulation Time point assessed Neutralizing antibody persistence Comparison to peak Ivan T Lee et al., 2023 BA.1-monovalent, BA.1-bivalent Month 3 post-booster Superior vs mRNA-1273 against BA.1 Maintained superiority A. Choi et al., 2021 Original mRNA-1273 6 months post-primary 6-7 fold decrease vs D614G; 24-69 fold decrease vs B.1.351/P.1 Peak at Month 1 K. Wu et al., 2021 mRNA-1273.351 6-8 months post-primary Low/undetectable before boost Waning evident A. Roederer et al., 2024 XBB.1.5 booster After multiple doses Improved persistence vs original Better against newer variants M. Echaide et al., 2023 Original formulations Up to 6 months Detectable but decreased Peak at 4 weeks A. Pegu et al., 2021 Original mRNA-1273 6 months Persisted but B.1.351 dropped considerably Variant-specific waning toof Pageof The BA.1-containing boosters maintained superior neutralizing antibody responses against Omicron BA.1 compared to the original vaccine through 3 months post-booster. With the original mRNA-1273 vaccine, neutralizing antibody titers showed a 6-7 fold decrease against wild-type D614G and a more dramatic 24-69 fold decrease against B.1.351 and P.1 variants over 6 months following the primary series. The rate of waning was particularly pronounced for the B.1.351 variant, with neutralizing antibodies dropping considerably by 6 months. Peak antibody responses occurred at 4 weeks post-vaccination for original formulations and remained detectable up to 6 months, albeit at reduced levels. Updated formulations showed improved persistence against their target variants. The XBB.1.5 booster demonstrated better maintenance of neutralization against newer variants compared to original formulations, with the first booster administered 8-9 months after primary vaccination and the second booster 4-6 months later. However, neutralization activity declined over time even with updated boosters, particularly against emerging variants like JN.1. ## Protection Against Breakthrough Infections Data on breakthrough infection rates with updated formulations compared to original vaccines were limited, with most studies focusing on immunogenicity rather than clinical outcomes. Aspect Findings Study population B.1.617.2 breakthrough Higher rates with low-dose formulations K18-hACE2 mice Disease severity Viral pneumonia with inflammation in breakthrough cases Preclinical models Dose-response relationship Low-dose (0.25 μg) showed breakthroughs vs high-dose (5 μg) 129S2 and K18-hACE2 mice Variant-specific protection Updated formulations offered better protection vs homologous strains Preclinical models Omicron protection Updated bivalent vaccines increased protection vs monovalent Clinical populations toof Pageof Preclinical studies revealed that breakthrough infections occurred primarily with the B.1.617.2 (Delta) variant, particularly when using low-dose vaccine formulations designed to model suboptimal immune responses. These breakthrough infections led to viral pneumonia with inflammation and airspace consolidation in K18-hACE2 mice. The effectiveness of variant-specific vaccines appeared superior against homologous strains, with updated formulations inducing greater antibody responses and conferring more protection against their target variants. Updated bivalent vaccines targeting Omicron variants offered increased protection against new Omicron subvariants compared to monovalent vaccines, though the original vaccines showed a significant decrease in protection against Omicron variants, with efficacy dropping to 30% after three doses. ## Synthesis The findings reveal systematic patterns in how updated mRNA-1273 formulations alter immunological mechanisms, with heterogeneity explained by specific variant targets, formulation strategy (monovalent vs bivalent), and the evolutionary timing of SARS-CoV-2 variants tested. ### Variant-Specific vs Broad Protection Trade-offs Monovalent variant-specific vaccines (mRNA-1273.351, mRNA-1273.529) consistently showed superior neutralization against their target variants—with geometric mean ratio increases of 1.67-1.77 for BA.1 formulations and twofold improvements for B.1.351 formulations—but demonstrated reduced neutralization against ancestral strains, with fourfold lower geometric mean titers against WA1/2020 D614G. This pattern reflects the antigenic distance between variant spike proteins and the immunodominant epitopes recognized by vaccine-induced antibodies. In contrast, bivalent formulations (mRNA-1273.211, mRNA-1273.214) maintained or improved ancestral strain neutralization while still boosting variant-specific responses. The BA.1-bivalent vaccine achieved a GMR of 1.11 against D614G, and mRNA-1273.211 showed a 46.4-fold increase against wild-type virus with GMT ratios ≥1 against all variants of concern. This breadth advantage comes at a modest cost: bivalent vaccines showed slightly lower peak titers against target variants (GMR 1.67) compared to monovalent formulations (GMR 1.77) for BA.1, suggesting antigenic competition or epitope masking when both ancestral and variant spike proteins are co-delivered. ### Dose-Dependent Breakthrough Risk The preclinical studies using high (5 μg) versus low (0.25 μg) doses revealed a critical threshold effect for protection. High-dose formulations of all vaccines—whether original mRNA-1273, variant-specific mRNA-1273.351, or bivalent mRNA-1273.211—conferred protection against weight loss and lung pathology across all tested variants including B.1.617.2 (Delta). However, low-dose formulations showed breakthrough lung infections and pneumonia specifically with B.1.617.2 in K18-hACE2 mice, despite producing approximately tenfold lower neutralizing activity. Notably, a minimum neutralizing titer of approximately 5000 was required to prevent lung infection, and low-dose vaccines fell below this threshold for certain variants. This dose-response relationship suggests that as immunity wanes over time in vaccinated populations—mimicking the reduced magnitude seen with low-dose vaccination—breakthrough infections become more likely, particularly with antigenically distant variants like B.1.617.2 that showed four-to-fivefold lower neutralizing responses across all vaccine types. The clinical relevance is supported by the observation that neutralizing antibodies declined 6-7 fold against D614G and 24-69 fold against B.1.351 and P.1 over 6 months following primary vaccination, potentially dropping below protective thresholds for some individuals and variants. ### Temporal Evolution and Escape Dynamics The longitudinal analysis across 2020-2024 revealed progressive viral escape from both original and updated vaccine-induced immunity. While many individual mutations emerging between 2020-2022 escaped sera from primary vaccination, few escaped boosted sera. However, newer variants demonstrated progressive loss of neutralization regardless of vaccine doses received. The bivalent booster improved neutralization against contemporaneous strains but not against JN.1, and the XBB.1.5 booster significantly increased titers against newer variants except JN.1. This pattern cannot be explained by simple waning immunity, as participants received up to 6 mRNA vaccines including updated formulations. Instead, the data suggest that SARS-CoV-2 evolution is outpacing vaccine updates. Each seasonal booster improves titers against contemporaneous strains, but novel variants continue to evade updated mRNA vaccines, demonstrating that antigenic drift occurs faster than the vaccine development-deployment cycle. The mechanistic basis likely involves progressive accumulation of mutations in immunodominant epitopes: the comprehensive testing panel of 131 mutations across 48 variants showed that newer variants like JN.1 carry combinations of mutations that collectively confer escape, even from updated boosters targeting XBB.1.5. ### Booster-Mediated Memory Recall The ability of updated formulation boosters to rescue waning immunity provides insight into B cell memory mechanisms. Six to eight months after primary vaccination, neutralizing antibodies against variants like B.1.351 and P.1 were low or undetectable. However, booster doses—whether with original mRNA-1273 or variant-modified formulations—increased neutralization titers to levels similar to or higher than peak titers after primary vaccination within two weeks, with statistically significant increases observed for mRNA-1273 and mRNA-1273.211 (P<0.0001). Importantly, the original mRNA-1273 booster induced a 16.7-fold increase against wild-type D614G, demonstrating robust homologous recall, while variant-modified boosters showed particularly strong responses: mRNA-1273.351 induced a 34.9-fold increase against B.1.351 and mRNA-1273.211 showed a 61.6-fold increase. These anamnestic responses indicative of robust B cell memory suggest that memory B cells formed during primary vaccination can be reactivated by booster antigens, even when those antigens differ antigenically from the priming strain. The magnitude of the response (up to 61.6-fold) implies affinity maturation and clonal expansion of memory cells, with cross-reactive memory B cells responding to shared epitopes between variants. This mechanism explains why both homologous and heterologous boosters effectively restore immunity, though variant-matched boosters achieve higher peak titers against their specific targets through preferential activation of variant-specific clones. ## References [A. Pegu, S. O'connell, S. Schmidt, S. O'Dell, C. A. Talana, and 27 more\\ (2021).Durability of mRNA-1273 vaccine–induced antibodies against SARS-CoV-2 variants. Science](/content/review/5ac7d462-1a70-4029-9382-e6d9f50ce22d/source/ss-236997251/index.html) [Kai Wu, A. Choi, M. Koch, Sayda M. Elbashir, LingZhi Ma, and 20 more\\ (2021).Variant SARS-CoV-2 mRNA vaccines confer broad neutralization as primary or booster series in mice. bioRxiv](/content/review/5ac7d462-1a70-4029-9382-e6d9f50ce22d/source/ss-233327536/index.html) [A. Choi, M. Koch, Kai Wu, L. Chu, LingZhi Ma, and 17 more\\ (2021).Safety and immunogenicity of SARS-CoV-2 variant mRNA vaccine boosters in healthy adults: an interim analysis. Nature Medicine](/content/review/5ac7d462-1a70-4029-9382-e6d9f50ce22d/source/ss-237536051/index.html) [M. Echaide, Luisa Chocarro de Erauso, A. Bocanegra, E. Blanco, G. Kochan, and 1 more\\ (2023).mRNA Vaccines against SARS-CoV-2: Advantages and Caveats. International Journal of Molecular Sciences](/content/review/5ac7d462-1a70-4029-9382-e6d9f50ce22d/source/ss-257714279/index.html) [K. Wu, A. Choi, M. Koch, L. Ma, A. Hill, and 15 more\\ (2021).Preliminary Analysis of Safety and Immunogenicity of a SARS-CoV-2 Variant Vaccine Booster. medRxiv](/content/review/5ac7d462-1a70-4029-9382-e6d9f50ce22d/source/ss-233767518/index.html) [M. Koch, Sayda M. Elbashir, Angela Woods, C. Henry, Charis Palandjian, and 14 more\\ (2021).Variant SARS-CoV-2 mRNA vaccines confer broad neutralization as primary or booster series in mice. Vaccine](/content/review/5ac7d462-1a70-4029-9382-e6d9f50ce22d/source/ss-243834267/index.html) [Baoling Ying, Bradley M. Whitener, L. VanBlargan, Ahmed O. Hassan, S. Shrihari, and 15 more\\ (2021).Protective activity of mRNA vaccines against ancestral and variant SARS-CoV-2 strains. Science Translational Medicine](/content/review/5ac7d462-1a70-4029-9382-e6d9f50ce22d/source/ss-244730216/index.html) [A. Roederer, Yi Cao, K. S. Denis, M. Sheehan, Chia Jung Li, and 8 more\\ (2024).Ongoing evolution of SARS-CoV-2 drives escape from mRNA vaccine-induced humoral immunity. medRxiv](/content/review/5ac7d462-1a70-4029-9382-e6d9f50ce22d/source/ss-268260010/index.html) [Baoling Ying, Bradley M. Whitener, L. VanBlargan, Ahmed O. Hassan, S. Shrihari, and 15 more\\ (2021).Protective activity of mRNA vaccines against ancestral and variant SARS-CoV-2 strains. bioRxiv](/content/review/5ac7d462-1a70-4029-9382-e6d9f50ce22d/source/ss-237349117/index.html) [Ivan T Lee, Catherine A Cosgrove, P. Moore, M. Bula, Philip A. Kalra, and 23 more\\ (2023).2363\. Three-Month Safety and Immunogenicity of Bivalent SARS-CoV-2 Omicron-Containing Booster Vaccines: Interim Results From a Phase 3, Randomized, Observer-Blind, Active-Controlled Trial. Open Forum Infectious Diseases](/content/review/5ac7d462-1a70-4029-9382-e6d9f50ce22d/source/ss-265459898/index.html) Download BIBDownload RISDownload TXT ## Report Status Gather sources 200 sources found Details Screen sources 10 sources included Details Extract data 80 data points extracted Details Generate report Save PDF BIBLaTeX, ZoteroRISZotero, MendeleyTXTAPA BibliographyPDFPDFDOCXMicrosoft Word Chat Got some follow-up questions? [Sign up](/content/users/auth?redirectToPath=%2Freview%2F5ac7d462-1a70-4029-9382-e6d9f50ce22d%2Fsource%2Fss-268260010&show=signup/index.html) or [sign in](/content/users/auth?redirectToPath=%2Freview%2F5ac7d462-1a70-4029-9382-e6d9f50ce22d%2Fsource%2Fss-268260010/index.html) to chat with this report. Back ## Ongoing evolution of SARS-CoV-2 drives escape from mRNA vaccine-induced humoral immunity A. Roederer, Yi Cao, K. S. Denis, M. Sheehan, Chia Jung Li, Evan C. Lam, David J. Gregory, M. Poznansky, John A. Iafrate, D. Canaday, S. Gravenstein, W. Garcia-Beltran, A. Balazs medRxiv· 2024· 18 citations SourceDOI Plain textPDF Searching for PDF Unable to find PDF from source Back Updated Formulation \- Specific vaccine name/code: Updated monovalent mRNA vaccine encoding the XBB.1.5 spike \- Variant(s) targeted: XBB.1.5 \- Type of formulation: Monovalent \- Dose amount: Not mentioned \- Difference from original mRNA-1273: Targets the XBB.1.5 variant Immunological Mechanisms \- Neutralizing antibody responses: Pseudovirus-based neutralization assays were used to measure the potency of sera from vaccine recipients. \- Binding antibody responses: Not mentioned. \- T-cell responses: Not assessed. \- Cellular immune responses: Not evaluated. \- Durability measures: Not explicitly discussed. \- Mechanistic pathway analysis: Not included. Variant Testing Panel \- Ancestral/original strain: Wuhan-Hu-1, D614G, WA1/2020 \- Specific variants of concern tested: Alpha, Beta (K417N, E484K, N501Y), Gamma, Delta (L452R, P681R), Omicron sublineages (BA.5, BQ.1.1, JN.1) \- Nomenclature used: WHO names, Pango lineages, spike mutations \- Testing was done against the variant the updated formulation was designed for vs other variants: Bivalent booster tested against BA.5 and earlier; XBB.1.5 booster tested against later variants with significant escape against Beta and JN.1 Comparative Mechanism Results \- Geometric mean titers/ratios with confidence intervals: Not mentioned \- Fold-change differences in antibody levels: Not mentioned \- Statistical significance of differences (p-values, confidence intervals): Not mentioned \- Superiority, non-inferiority, or inferiority results: Updated formulations (bivalent and XBB.1.5 boosters) show superiority in breadth of immune response but inferiority in depth against certain variants (e.g., JN.1). \- Cross-variant protection differences: Updated formulations improve protection against newer variants but with exceptions (e.g., JN.1). \- Breadth vs depth of immune responses: Boosting enhances breadth but with limitations in depth against certain variants. \- Direction of effect (better, worse, equivalent) for each mechanism: Generally better for breadth, worse for depth against certain variants. Vaccination Regimen \- Primary vaccination series: 2 shots \- Booster dose administration: 3 shots (first booster 8-9 months after primary series, second booster 4-6 months after first booster) \- Number of prior doses before updated formulation: Up to 6 mRNA vaccines \- Timing between doses: First booster 8-9 months after primary series, second booster 4-6 months after first booster \- Previous vaccine types received: Homologous (Pfizer or Moderna) and heterologous (mix of Pfizer and Moderna) \- Concurrent administration with other vaccines: Not mentioned \- Schedule used: 2-dose primary, 3rd dose booster, 4th dose, etc. Breakthrough Protection \- Breakthrough infection rates by variant: Not mentioned \- Disease severity in breakthrough cases: Not mentioned \- Viral load differences during breakthrough: Not mentioned \- Protection against symptomatic vs asymptomatic infection: Not mentioned \- Effectiveness estimates comparing updated vs original formulations: Updated XBB.1.5 booster increases titers against newer variants, but novel variants continue to evade updated mRNA vaccines. \- Duration of protection differences: Not mentioned Study Population \- Age groups studied: Adult (median age 33), Elderly (median age 72) \- Prior SARS-CoV-2 infection status: COVID-naive donors \- Immunocompromised status: Not mentioned \- Time since last vaccination or infection: Not explicitly mentioned \- Geographic location: Not specified \- Sample size for mechanism analyses: 20 serum donors for primary vaccine series; 29 nursing home residents and 7 healthcare workers for longitudinal analyses Mechanism Durability \- Time points measured: First booster dose 8-9 months after primary vaccination, second booster 4-6 months after first booster. \- Rate of antibody waning over time: Decline in neutralization activity against newer variants over time, especially for WT and bivalent boosters. \- Persistence differences between updated vs original formulations: Updated formulations (XBB.1.5 booster) showed improved persistence compared to original formulations. \- Memory response durability: Increased breadth of neutralizing activity after boosting. \- Peak responses occurred: After booster vaccinations. \- Mechanism differences maintained over time: Yes, updated boosters continued to show improved neutralization against newer variants. Since the COVID-19 pandemic began in 2020, viral sequencing has documented 131 individual mutations in the viral spike protein across 48 named variants. To determine the ability of vaccine-mediated humoral immunity to keep pace with continued SARS-CoV-2 evolution, we assessed the neutralization potency of sera from 76 vaccine recipients collected after 2 to 6 immunizations against a comprehensive panel of mutations observed during the pandemic. Remarkably, while many individual mutations that emerged between 2020 and 2022 exhibit escape from sera following primary vaccination, few escape boosted sera. However, progressive loss of neutralization was observed across newer variants, irrespective of vaccine doses. Importantly, an updated XBB.1.5 booster significantly increased titers against newer variants but not JN.1. These findings demonstrate that seasonal boosters improve titers against contemporaneous strains, but novel variants continue to evade updated mRNA vaccines, demonstrating the need for novel approaches to adequately control SARS-CoV-2 transmission. INTRODUCTION Since it was first described in late 2019, SARS-CoV-2 has undergone continuous evolution. The virus originated in Wuhan, China, and was declared a global pandemic on March 11, 2020 by the WHO 1,2 . Since then, it has infected over 700 million people and caused nearly 7 million deaths 2,3 . The first single mutation to become fixed in the population, D614G, emerged in February of 2020 and was shown to be more infectious and stable than the original sequence 4,5 . However, the first variant to be named, B.1.1.7 or Alpha, harbored multiple mutations and became the dominant strain in the UK, exhibiting over 50% greater infectivity than wild-type (WT) 6 . As the pandemic entered its second year, additional variants emerged carrying mutations in the receptor binding domain (RBD) that enhanced transmissibility 7,8 through improved ACE2 binding and exhibited viral escape from convalescent sera \[9\]\[10\]\[11\] . In the subsequent two years, the number of variants that have been documented has increased significantly, with new mutations (predominantly in the RBD) that enable escape from vaccine-induced neutralizing antibodies and enhance transmissibility 10,\[12\]\[13\]\[14\]\[15\] . An unprecedented effort to develop effective countermeasures resulted in multiple vaccines being approved by the FDA in the United States, all of which were based upon the WT SARS-CoV-2 spike protein containing stabilizing mutations that were previously identified for RSV and MERS 16,17 . Numerous vaccine technologies were employed, including mRNA containing lipid nanoparticles, recombinant proteins, or adenovirus vectored vaccines \[18\]\[19\]\[20\]\[21\] , however mRNA vaccines were the most widely deployed in the United States. Clinical trials of these vaccines demonstrated remarkable efficacy to reduce COVID-19 infections, hospitalizations, and deaths \[18\]\[19\]\[20\]\[21\] . However, the emergence of the Omicron variant and declining antibody titers over time led to the recommendation of a third booster dose of mRNA vaccine by the FDA, yielding neutralizing, cross-reactive antibodies against Omicron 13,22 . The emergence of new variants such as BA.5 in 2022, necessitated the reformulation of vaccines 23 , spurring the transition from monovalent to bivalent vaccines incorporating both WT and BA.5 spikes 24 . Despite the efficacy of the original vaccines in preventing severe disease and death, their effectiveness waned against heavily mutated, highly transmissible variants such as BQ.1.1, XBB, BA.2.86 and their sub-variants 15,\[25\]\[26\]\[27\] . These highly evolved strains carried 34-58 total mutations within the spike protein, with 28 of these mutations in the RBD in the most recent circulating variant JN.1. Recently, an FDA advisory committee recommended deployment of an updated monovalent mRNA vaccine encoding the XBB.1.5 spike 28 . However, a variant named BA.2.86 has arisen recently that harbors 57 total spike mutations, including 20 that have not been reported previously, raising concerns that derivatives of this strain could escape the XBB.1.5 booster. BA.2.86 and its derivatives including JN.1 now make up over 90% of new COVID sequences being submitted to GISAID 29,30 . To understand the impact of mutations which have arisen across the shifting vaccination landscape, we conducted a comprehensive analysis of individual mutations found in variants of concern or interest which emerged during the pandemic. Using a previously validated high throughput neutralization assay 10,13,31 , we measured the neutralization of pseudoviruses representing each individual mutation across study participants who had received either the primary vaccine series (two shots) or the booster series (three shots). Furthermore, we evaluated the neutralization activity of sera from boosted individuals against all individual mutations, as well as strains harboring combinations of mutations through JN.1, totaling 220 different pseudoviruses. Interestingly, we find that a handful of individual mutations arising after Omicron enable significant escape from boosted serum. Remarkably, the most recent variants, including BQ.1.1, XBB, and XBB.1.5, exhibit escape from boosted sera that is comparable to SARS-CoV-1 and the related pre-emergent WIV1 strain. However, sera from individuals receiving bivalent boosters exhibited significant neutralizing activity against more recent strains. Despite this, more recent variants, such as EG.5. 1 and HK.3, exhibit substantial escape from bivalent boosted sera, suggesting that SARS-CoV-2 remained ahead of efforts to update vaccines. We find that the latest XBB.1.5 booster enhances neutralizing activity against many variants that escaped bivalently boosted sera, however the latest JN.1 variant remains significantly resistant to neutralization across vaccinations. Together, these results highlight the importance of continued surveillance of SARS-CoV-2 sequence evolution and support the updating of existing vaccines. At the same time, our results also highlight the need for novel approaches capable of inducing broadly neutralizing humoral immunity to counter the continued evolution of SARS-CoV-2 variants. SARS-CoV-2 variants of concern have accumulated spike mutations focused within RBD SARS-CoV-2 has resulted in over 700 million infections and nearly 7 million deaths since it first emerged in 2020 2,3 . The virus, which originated in Wuhan China, circulated for nearly a year before the first variant of concern (Alpha) emerged (Figure 1A , Table S1 ), however in subsequent years, numerous variants of concern and interest have been described (Figure 1B , Table S1 ). One of the most globally dominant variants, Delta (B.1.617.2), was ascendant in June 2021 and harbored 8 mutations, including 2 in the RBD 6 . However, the Omicron variant (BA.1), containing 32 spike mutations, rapidly rose to prominence after it first appeared in November 2021. Variants derived from BA.1 continued to emerge resulting in the BA.5 strain harboring two additional mutations within the RBD. Most recently, JN.1 has become the dominant strain globally 30 , harboring 58 total spike mutations with 27 of these in the RBD 29 . Impressively, the SARS-CoV-2 pandemic has now resulted in variants that are further evolved from the original Wuhan strain than other distinct coronaviruses such as RaTG13 32,33 (Figure S1A ) which harbors only 26 amino acid differences from the Wuhan strain. Over the course of the pandemic, 131 differences in viral spike have arisen across variants of concern/interest, including several which occurred at the same position in the spike protein (Figure S1B ). Neutralization assays with primary vaccine sera reveals regions of vulnerability within the natural mutation landscape To investigate the effect of all individual spike variations on vaccine induced humoral immunity we studied a cohort of 20 serum donors who had received the primary vaccine series (two shots of Pfizer (BNT162b)). The cohort had a median age of 33 and was 50% female. Sera from this group was subjected to pseudovirus-based neutralization assays that we and others have previously validated 10,13,31,\[34\]\[35\]\[36\]\[37\]\[38\] . In brief, pseudotyped lentiviral particles bearing a given SARS-CoV-2 spike protein and encoding a luciferase-expressing transgene were produced and mixed with serially diluted donor serum before ACE2-expressing target cells were added (Figure 2A ). Assays were performed using pseudoviruses representing 69 differences present across 24 named variants of concern/interest up-to and including Delta. Variants with multiple consensus sequences such as Beta (V1-V6) are denoted as separate variants (Figure S2 ). Donor serum exhibited high neutralizing activity against the Wuhan pseudovirus, with a geometric mean pseudovirus 50% neutralization titer (pNT 50 ) of 2962 (Figure 2B ). However, individual changes in either the N-terminal domain (NTD), RBD or portions of the S1 region exhibited significant escape from neutralization. Within the RBD in particular, K417N, K417T, L452R, and Y453F had highly significant reductions (P<0.0001) in neutralization titers with geometric means of between 300-500. Other mutations in the NTD and S1 subunit significantly reduced pNT 50 including P26S, T29I, Q52R, D80A, T572I, Q677H, and P681R while those found within the S2 domain had little effect on neutralization titers except V1176F and M1229I. To explore patterns of individual vaccine responses, two-way hierarchical clustering was performed across all mutations tested for each individual donor (Figure 2C ). This analysis revealed that individual sera had remarkably similar loss of neutralization activity across a subset of mutations within the RBD, NTD, and S1 with more than half of all mutations being harder to neutralize than WT. Interestingly, two separate clusters of donors were observed in this cohort, with the bottom cluster having a broader overall response to individual mutations arising through the Delta strain. However, we found no correlation across time, age or sex to account for vaccination outcomes (not shown). These results suggest that humoral responses in most individuals were focused on particular epitopes within the NTD, RBD and S1, following the primary vaccination regimen with approximately half of these individuals generating broader responses. Combinations of individual mutations contribute to neutralization resistance Given that variants of concern/interest harbor multiple mutations, we assessed the impact of each individual mutation in the spike of the Beta variant. Of note, no individual mutation could account for the differences in pNT 50 between WT and Beta variant spike, but K417N and E484K were both highly neutralization resistant (Figure 3A , left). We previously showed that the RBD region of Beta was largely responsible for escape from neutralization 10 . To understand how multiple RBD mutations might coordinate to escape humoral immunity, primary vaccine sera was tested against double and triple RBD mutants (Figure 3A , right). Neither single mutation came close to recapitulating the neutralization resistance of the parent spike, but we did observe an additive effect for the three individual RBD mutations found in the Beta variant. Combinations of any two mutations yielded greater resistance to neutralization than single mutations, but the triple mutant exhibited nearly as much escape as the full Beta spike. Despite the neutralization resistance of Beta variant pseudovirus, genomic surveillance suggests that the Delta variant rapidly overtook all other circulating SARS-CoV-2 variants over a period of eight months 30,39,40 . Primary vaccine sera was also assessed against each of the individual Delta variant mutations, with L452R and P681R exhibiting significant reduction of pNT 50 (Figure 3B , left). In contrast to Beta, combining RBD mutations in Delta resulted in no additional reduction in pNT 50 , suggesting that mutations outside of RBD, namely P681R, appear to coordinate with L452R to drive greater escape (Figure 3B , right). Similar analyses were performed for other designated variants of concern (VOC) and their individual spike mutations including the Alpha (Figure S3A ), Gamma (Figure S3B ), and Epsilon variants (Figure S3C ). Among these, all variant spikes exhibited reduced neutralization by sera from primary vaccination, likely from individual mutations within S1 for Alpha, RBD and NTD for Gamma and the RBD for Epsilon. Taken together, our results indicate that multiple mutations often coordinated to reduce neutralizing activity against variants of concern following primary vaccination. mRNA booster vaccination significantly enhances neutralization of SARS-CoV-2 mutants During a significant spike in COVID-19 cases in July 2021 stemming from the Delta variant, a third mRNA booster shot consisting of the same Wuhan spike formulation as the primary series was introduced in the United States 41 (Figure 1A ). To measure the impact of a third mRNA administration on humoral immunity, 20-22 samples from primary vaccinee or mRNA boosted donors were matched for age and subjected to neutralization assays across the same set of individual mutations reported through October 2021 (up-to the Delta variant) (Figure S4A, S5 ). The vast majority of mutants tested were more potently neutralized relative to the WT titer by serum from boosted donors than those who had only received the primary series (Figure 4A ). Two-way hierarchical clustering was performed across all mutations tested for each individual donor (Figure 4B ). Boosting greatly enhanced the breadth of neutralizing activity relative to WT titer for all except one donor (Figure 4B ). Across nearly all variants tested, there was improved neutralization potency relative to wildtype, suggesting an overall broadening of the humoral response (Figure 4C ). Interestingly, some mutants, including D80A, K417N, T572I, and Q677H, exhibited disproportionately greater neutralization, suggesting that these epitopes were preferentially targeted by booster vaccination (Figure 4C ). Comparison between the two groups found that donors who received the booster vaccine had significantly increased neutralization activity overall (Figure 4D ). Comparing neutralization activity across mutations positioned within different spike regions showed significantly increased activity in boosted samples across spike regions, including NTD (Figure S4B ), S1 (Figure S4C ), RBD (Figure S4D ) and S2 (Figure S4E ). Recent variants reveal progressively more escaped mutations from boosted sera In November of 2021, a SARS-CoV-2 variant harboring an unusually large number of mutations was first identified in South Africa 42 . This novel variant of concern, designated Omicron (BA.1), contained 59 mutations in its genome, 32 of which were in the spike protein. The heavily mutated spike contained 15 RBD mutations, raising the possibility that it had escaped from humoral immunity. A number of heavily mutated, highly transmissible variants have descended from BA.1 since this time 25,43,44 . Additional neutralization assays were performed to assess whether boosted sera exhibited improved activity against more recent mutations (Figure 5A , Figure S6 ). While sera from individuals receiving a third mRNA booster vaccine was able to neutralize most mutants, spike mutations V445P, N460K and F486P exhibited 8.6-, 6.5-, and 7.7-fold decrease in pNT 50 respectively (Figure 5A ). Across the full set of mutations since the beginning of the pandemic, a number of individual mutants in the RBD escaped from mRNA-boosted serum samples (Figure S6 , Figure S7A ). Compared to mutations that had emerged in earlier variants including Delta, those emerging in Omicron and subsequent variants demonstrated significantly less neutralization by boosted sera (P = 0.0021) (Figure 5B ). Mapping these mutations onto the structure of the SARS-CoV-2 spike showed that mutations appearing after Delta were largely concentrated at the apex of the spike and specific residues within the NTD (Figure 5C ). Mapping the neutralization resistance of all 131 mutations onto the spike protein crystal structure revealed that the most escaped mutants appeared at the apex of the spike and on the outer edges of the NTD (Figure S7B ). Neutralization values against RBD mutants over time revealed a significant reduction in pNT 50 relative to WT (P = 0.02), suggesting that more recent mutations drove escape from vaccine responses (Figure 5D ). Analysis of boosted neutralizing titers against mutations over time showed an overall trend towards greater escape (R 2 = 0.08, P = 0.0006), but relatively few novel mutations have been described in variants of concern/interest between 2022 and 2023 (Figure S7C ). Neutralization of more recent strains showed a 5-fold decrease in pNT 50 value beginning with BA.1, further decreasing to 104-fold for the most escaped among them, BQ.1.1.22 (Figure 5E ). Two-way hierarchical clustering was performed across all mutations through the XBB.1.16 variant for boosted neutralizing titers (Figure S7D ). Mutations that clustered together that were hardest to neutralize are all present in variants that descended from XBB. Donors separated into two clusters, one with a broader overall response and another cluster with notable weaknesses against mutations present in XBB variants. Infectivity variants During a surge in cases, largely attributed to the BA.5 variant, bivalent mRNA vaccines encoding both WT and BA.5 spike sequences were approved for use in the United States in August of 2022 23,45 . With the recombination of spikes producing XBB and sub variants, and the waning immunity of bivalent vaccination against these variants 46,47 , a newly reformulated vaccine encoding only the XBB.1.5 spike sequence was recommended for use in September 2023 48,49 . We obtained longitudinal samples from 29 nursing home residents (median of 72 years old) and 7 health care workers (median age 60) who received either three or four WT mRNA vaccines followed by a bivalent booster and an XBB.1.5 booster. To measure the impact of each vaccine administration, longitudinal serum samples were tested against variants previously found to escape boosted sera, or which were dominant variants for more than a month (Figure 6A ). While there was a clear trend of increased neutralizing activity with each successive vaccination, only modest differences were observed between the 3rd and 4th WT mRNA boost (Figure 6B ). Bivalent vaccines improved pseudovirus neutralization titers after BA.5, though the activity against these variants was still significantly lower than WT (Figure S8A ). In contrast, only the most recent dominant variant, JN.1, and the original Beta variant showed significant escape from the XBB.1.5 boosted samples (Figure S8B ). Overall, we observed increasing neutralization titers against all pseudoviruses tested relative to WT in the same cohort across each successive vaccination from WT booster to bivalent booster, and finally to XBB.1.5 booster (Figure 6C ). We also observed a significant decline in neutralization activity across variants that arose across time for WT and Bivalent boosters but no significant decline across variants following XBB.1.5 booster vaccination (Figure 6D ). Two-way hierarchical clustering was performed on matched donors at the boosted time point to understand the effect of additional boosters against new vaccinations (Figure 6E ). Donors separated into two clusters after receiving a 3rd WT booster; one with broader neutralizing activity and another with lower activity against BQ.1.1.22 and later strains. However, the distinction between these two clusters was largely attenuated with each successive administration of Bivalent and XBB1.5 boosters. Given the vast number of mutations in new variants, we wanted to assess their effect on the efficiency of viral entry. To determine the viral entry efficiency of variant spikes, we compared the rate of pseudovirus transduction over a defined range of concentrations (Figure S8C , S8D). Remarkably, spikes from all variants arising after BA.1 were substantially better at entering cells than WT, Beta, or Delta. When compared to WT, viral entry of Beta, Delta, BA.1, and BA.2 were all slightly enhanced, whereas variants from BA.5 to HK.3 infected 10-30 times more efficiently than WT (Figure 6F ). Notably, BA.2.86 (which is derived from BA.2) was similar in its ability to enter cells as BA.2, but its descendant JN.1 has become the dominant circulating strain in the United States, with 15-fold greater efficiency at cell entry than WT. Taken together, these observations suggest that Bivalent and XBB.1.5 boosters substantially improve the neutralization activity of sera against highly mutated variants, but that there are substantial differences in the efficiency of spike mediated cell entry which has significantly increased in more recent strains. DISCUSSION Although the COVID-19 pandemic declaration was ended by the WHO in May 2023, SARS-CoV-2 continues to transmit globally in the form of novel variants. We sought to understand how the constellation of mutations that have appeared since the beginning of the pandemic were impacted by the changing landscape of humoral immunity driven by updated vaccines. To this end, we determined the neutralization potency of patient sera against pseudoviruses representing each of the individual mutations observed within 50 strains of SARS-CoV-2 that have emerged, totaling 220 pseudoviruses. Our study utilized samples from 20-22 COVID naive donors who received a primary vaccination series, defined as either two shots of the Pfizer 18 or Moderna 19 mRNA vaccines, or a third mRNA booster dose. We also analyzed longitudinal samples from an older cohort consisting primarily of nursing home residents, taken after each successive vaccination. These individuals received up to a total of six mRNA vaccines, including the recent XBB.1.5 booster. We find that sera from donors who received a primary vaccine series were especially vulnerable to individual mutations across multiple regions of the spike protein, including NTD, RBD, S1 and the C-terminus of S2. All individual mutations within the RBD resulted in escape from primary vaccination. Specifically, three receptor-binding domain (RBD) mutations from the Beta variant (K417N, E484K, N501Y) and two mutations from the Delta variant (L452R, P681R) substantially lowered the neutralization activity of sera from primary donors. Interestingly, sera from donors who received a third mRNA booster dose exhibited significantly improved neutralization across many, but not all, of the individual mutants we tested. New clusters of mutations formed compared to primary series clustering; notably, P681R and L452R clustered together and were among the hardest single mutants to neutralize, suggesting that these mutations were critical to the Delta variant overtaking other VOCs at the time. We observed that boosting disproportionately increased titers against specific key escape mutations across the spike including D80A, K417N, T572I and Q677H. Consistent with our previous report 13 , we found that boosting resulted in substantially greater breadth of neutralization against more recent strains. However, we observed a distinct loss of activity for boosted sera beginning with the emergence of the BQ.1.1 variant. Interestingly, this appeared to be largely driven by novel RBD mutations at positions 445 and 486, which have been retained in all variants since XBB, suggesting that they provide a significant advantage to the virus. In response to the rapid emergence of SARS-CoV-2 variants, the vaccine formulation was twice updated; first to deliver a bivalent immunogen containing spike proteins from the WT and BA.4/BA.5 strain 50,51 and then again to deliver a monovalent XBB.1.5 spike sequence 48,52 . We find that the bivalent booster results in neutralizing activity comparable to wildtype for variants BA. 5 and earlier, while also improving activity against BQ.1.1 and later variants, although these later variants still significantly escaped bivalent booster vaccination. In contrast, we find that the XBB.1.5 booster exhibits substantial improvement against later variants and only saw significant escape against the prior Beta variant and currently dominant JN.1 variant. In addition to escape from vaccine sera, we found that mutations also contribute significantly to the ability of pseudovirus to infect cells, suggesting that variant selection is optimizing both antibody escape and viral entry. We found that spikes from variants post BA.1 produced pseudoviruses that were up to 30-fold better at transducing target cells than wild-type, suggesting that WT SARS-CoV-2 spike was not optimally configured for ACE2-dependent viral entry. Our study demonstrates that individuals receiving the full mRNA vaccination regimen, including the XBB.1.5 booster, have maximally effective neutralizing activity against recent strains of SARS-CoV-2. However, despite improvements mediated by XBB.1.5 boosters against more recently emerging strains, there is still escape, particularly by the latest BA.2.86 descendant JN.1. This highlights the ongoing challenge presented by continued evolution of SARS-CoV-2, which has largely outpaced attempts to update vaccines. Our findings support the development of novel vaccine and prevention modalities capable of eliciting broad protection against future variants/outbreaks. Of note, a number of monoclonal antibodies have been described that target highly conserved epitopes such as the stem helix \[53\]\[54\]\[55\]\[56\] or fusion peptide 57- 60 , which are capable of neutralizing across the entire Coronaviridae family. Next-generation vaccines that exploit these and other sites of vulnerability across Coronaviruses will likely be necessary to combat the ongoing evolution of SARS-CoV-2 and may help prevent future pandemics. Limitations of the study While numerous prior studies have shown that neutralization activity against pseudoviruses is well-correlated to replication competent SARS-CoV-2 \[34\]\[35\]\[36\]\[37\]\[38\] , it is possible that the mutations in the more recent variant spike proteins may cause them to behave differently than previously tested variants. In addition, while we confirmed that ACE2 expression is required for infection of 293T cells, natural target cells in the respiratory tract may express alternative receptors or attachment factors that facilitate infection and are not adequately modeled in our system. The study is limited to the evaluation of serum neutralizing antibodies to the spike protein for our pseudovirus neutralization assay, and does not take into account neutralizing antibodies to other proteins including the N and M proteins. Additionally, our infectivity studies only take into account the infectivity of spike mutations, not mutations in other proteins that contribute to viral infectivity. Furthermore, we did not assess other antibodymediated functions such as complement deposition, antibody-dependent cellular cytotoxicity, or antibody-dependent cellular phagocytosis, which may contribute to protection even in the absence of neutralizing antibodies. We also did not assess the role of vaccine-elicited cellular immune responses mediated by T cells and NK cells, which are likely to play a key role in disease prevention for vaccine recipients. Mutations highlighted for each strain are from corresponding table S1 . Lead Contact Further information and requests for resources and reagents should be directed to and will be fulfilled by Alejandro Balazs (abalazs@mgh.harvard.edu). Materials Availability Plasmids generated in this study will be available through Addgene. Recombinant proteins and antibodies are available from their respective sources. Data and Code Availability This study did not generate sequence data. Data generated in the current study (including neutralization measurements and flow cytometric files) have not been deposited in a public repository but are available from the corresponding author upon request. The code used to make the phylogenetic trees can be made available from the corresponding author upon request. Participants received their first booster dose 8-9 months after the primary vaccination series, and their second booster 4 to 6 months after the first booster. We report results from blood samples obtained at time points prior to and following the two booster doses. Cell lines HEK 293T cells (ATCC) were cultured in DMEM (Corning) containing 10% fetal bovine serum (VWR), and penicillin/streptomycin (Corning) at 37°C/5% CO 2 . 293T-ACE2 cells were a gift from Michael Farzan (Scripps Florida) and Nir Hacohen (Broad Institute) and were cultured under the same conditions. Confirmation of ACE2 expression in 293T-ACE2 cells was done via flow cytometry. Construction of variant spike expression plasmids To create variant spike expression plasmids, we performed multiple PCR fragment amplifications utilizing oligonucleotides containing each desired mutation (Integrated DNA Technology) and utilized overlapping fragment assembly to generate the full complement of mutations for each strain. Importantly we generate these mutations in the context of our previously described codon-optimized SARS-CoV-2 spike expression plasmid harboring a deletion of the C-terminal 18 amino acids that we previously demonstrated to result in higher pseudovirus titers. Assembled fragments were inserted into NotI/XbaI digested pTwist-CMV-BetaGlobin-WPRE-Neo vector utilizing the In-Fusion HD Cloning Kit (Takara). All resulting plasmid DNA utilized in the study was verified by whole-plasmid deep sequencing (Illumina or Primordium Labs) to confirm the presence of only the intended mutations. SARS-CoV-2 pseudovirus neutralization assay To compare the neutralizing activity of vaccinee sera against coronaviruses, we produced lentiviral particles pseudotyped with different spike proteins as previously described (Garcia-Beltran et al. 2021). Briefly, pseudoviruses were produced in 293T cells by PEI transfection of a lentiviral backbone encoding CMV-Luciferase-IRES-ZsGreen as well as lentiviral helper plasmids and each spike variant expression plasmid. Following collection and filtering, production was quantified by titering via flow cytometry on 293T-ACE2 cells. Neutralization assays and readout were performed on a Fluent Automated Workstation (Tecan) liquid handler using 384-well plates (Grenier). Three-fold serial dilutions ranging from 1:12 to 1:8,748 were performed for each serum sample before adding 125-250 infectious units of pseudovirus for 1 h. Subsequently, 293T-ACE2 cells containing polybrene were added to each well and incubated at 37°C/5% CO 2 for 48-60 hrs. Following transduction, cells were lysed using a luciferincontaining buffer (Siebring-van Olst et al. 2013) and shaken for 5 min prior to quantitation of luciferase expression within 1 h of buffer addition using a Spectramax L luminometer (Molecular Devices). Percent neutralization was determined by subtracting background luminescence measured in cell control wells (cells only) from sample wells and dividing by virus control wells (virus and cells only). Data was analyzed using Graphpad Prism and pNT 50 values were calculated by taking the inverse of the 50% inhibitory concentration value for all samples with a neutralization value of 80% or higher at the highest concentration of serum. Titering To determine the infectious units of pseudotyped lentiviral vectors, we plated 400,000 293T-ACE2 cells per well of a 12-well plate. 24 h later, three ten-fold serial dilutions of neat pseudovirus supernatant were made in 100 μL, which was then used to replace 100 μL of media on the plated cells. Cells were incubated for 48 h at 37°C/5% CO 2 to allow for expression of ZsGreen reporter gene and harvested with Trypsin-EDTA (Corning). Cells were resuspended in PBS supplemented with 2% FBS (PBS+), and analyzed on a Stratedigm S1300Exi Flow Cytometer to determine the percentage of ZsGreen-expressing cells. Infectious units were calculated by determining the percentage of infected cells in wells exhibiting linear decreases in transduction and multiplying by the average number of cells per well determined at the initiation of the assay. At low MOI, each transduced ZsGreen cell was assumed to represent a single infectious unit. Quantitation of pseudovirus by RT-qPCR To determine the genome copy concentration of pseudotyped lentiviral vectors, lentiviral RNA was extracted from pseudovirus supernatant using the QIAamp viral RNA mini kit (Qiagen). Each sample was serially diluted, and each dilution was treated with 1.2 U of Turbo DNase (Invitrogen) at 37°C for 30 min followed by heat inactivation at 75°C for 15 min. 10 μL of the treated RNA was used in a 20 μL qRT-PCR reaction with the qScript XLT one-step RT-qPCR Tough Mix, low ROX mix (Quanta Biosciences), a TaqMan probe containing locked nucleic acids (/56-FAM/AGC+C/i5NitInd/GG+GA/ZEN/GCTCTCTGGC/3IABkFQ/) (IDT), and primers designed for targeting the LTR gene of NL4-3 HIV genome, from which the lentiviral vector was derived (5′-GGTCTCTCTIGITAGACCAG and 3′-TTTATTGAGGCTTAAGCAGTGGG). Each dilution was run in duplicate on a QuantStudio 12K Flex (Applied Biosystems). The following cycling conditions were used: 50°C for 10 min, 95°C for 3 min followed by 50 cycles of 95°C for 15 s and 60°C for 1 min. Virus titer was determined by comparison with a standard curve generated using a plasmid standard generated from serial dilution of CMV-Luciferase-IRES-ZsGreen lentiviral backbone. DNase and No DNase controls were also included at 2.5 × 10 8 GC/mL of the same plasmid. The range of the assay was from 2.5 × 10 7 GC/mL to 1.5 × 10 3 GC/mL. Upon analysis, the average of the three most concentrated dilutions within range of the standard were used to calculate genome copies/mL. Mutations highlighted for each strain are from corresponding table S1 . Mutations where a significant majority of donor titers reached the limit of detection were also excluded. Clustering was performed using pheatmap package v1.0.12 in R-Studio. A (wild type) DECLARATIONS OF INTEREST A.B.B. is a founder of Cure Systems LLC. INCLUSION AND DIVERSITY We worked to ensure gender balance in the recruitment of human subjects. We worked to ensure ethnic or other types of diversity in the recruitment of human subjects. We worked to ensure that the study questionnaires were prepared in an inclusive way. One or more of the authors of this paper self-identifies as an underrepresented ethnic minority in science. One or more of the authors of this paper self-identifies as a member of the LGBTQ+ community. Supplemental Figure 6 : All mutations tested against boosted sera. All mutations and variants tested against 24 mRNA boosted donor samples. Acknowledgements ACKNOWLEDGEMENTS, FUNDING SUPPORTWe wish to thank Michael Farzan, PhD, for providing ACE2-expressing 293T cells.This work was supported by the Peter and Ann Lambertus Family Foundation.A.B.B. was supported by NIAID R01s AI174875, AI174276, the NIDA Avenir New Innovator Award DP2DA040254 a Massachusetts Consortium on Pathogenesis Readiness (MassCPR) grant and CDC subcontract 200-2016-91773-T.O.2. Failed to load PDF: Paper sources Abstract screening pilot Abstract screening results Extraction pilot Extraction results Research report * * * Modify setup StripeM-Inner