key: cord-1000513-oqz6i77n authors: Payne, Rebecca P.; Longet, Stephanie; Austin, James A.; Skelly, Donal T.; Dejnirattisai, Wanwisa; Adele, Sandra; Meardon, Naomi; Faustini, Sian; Al-Taei, Saly; Moore, Shona C.; Tipton, Tom; Hering, Luisa M.; Angyal, Adrienn; Brown, Rebecca; Nicols, Alexander R.; Gillson, Natalie; Dobson, Susan L.; Amini, Ali; Supasa, Piyada; Cross, Andrew; Bridges-Webb, Alice; Reyes, Laura Silva; Linder, Aline; Sandhar, Gurjinder; Kilby, Jonathan A.; Tyerman, Jessica K.; Altmann, Thomas; Hornsby, Hailey; Whitham, Rachel; Phillips, Eloise; Malone, Tom; Hargreaves, Alexander; Shields, Adrian; Saei, Ayoub; Foulkes, Sarah; Stafford, Lizzie; Johnson, Sile; Wootton, Daniel G.; Conlon, Christopher P.; Jeffery, Katie; Matthews, Philippa C.; Frater, John; Deeks, Alexandra S.; Pollard, Andrew J.; Brown, Anthony; Rowland-Jones, Sarah L.; Mongkolsapaya, Juthathip; Barnes, Eleanor; Hopkins, Susan; Hall, Victoria; Dold, Christina; Duncan, Christopher JA.; Richter, Alex; Carroll, Miles; Screaton, Gavin; de Silva, Thushan I.; Turtle, Lance; Klenerman, Paul; Dunachie, Susanna title: Immunogenicity of standard and extended dosing intervals of BNT162b2 mRNA vaccine date: 2021-10-16 journal: Cell DOI: 10.1016/j.cell.2021.10.011 sha: 50cfd1adb8855388aeb650fb98886d6f990a9ead doc_id: 1000513 cord_uid: oqz6i77n Extension of the interval between vaccine doses for the BNT162b2 mRNA vaccine was introduced in the UK to accelerate population coverage with a single dose. At this time, trial data was lacking, and we addressed this in a study of UK healthcare workers. The first vaccine dose induced protection from infection from the circulating alpha (B.1.1.7) variant over several weeks. In a sub-study of 589 individuals, we show that this single dose induces SARS-CoV-2 neutralizing antibody (NAb) responses and a sustained B and T cell response to spike protein. NAb levels were higher after the extended dosing interval (6-14 weeks) compared to the conventional 3-4 week regimen, accompanied by enrichment of CD4+ T cells expressing IL2. Prior SARS-CoV-2 infection amplified and accelerated the response. These data on dynamic cellular and humoral responses indicate that extension of the dosing interval is an effective, immunogenic protocol. The vaccine dosing interval was either the conventional "short" 2-5 week interval (n=86, median 24 176 days, IQR 21-27, range 14-35), or a "long" 6-14 week interval (n=503, median 71 days, IQR 64-77, 177 range 45-105). ( Figure 1B and Table 1 ). An overview of assays is in Table S1 . Priming and boosting of serologic responses to SARS-CoV-2 using the extended dosing schedule 180 We next explored the immune responses accompanying this protection in the inter-dose interval. Firstly, using SARS-CoV-2 naïve individuals in the smaller immunology cohort (PITCH), NAb levels 182 were tested using a live virus micro-neutralization assay as previously reported (Dejnirattisai et al., 183 2021; Liu et al., 2021; Supasa et al., 2021; Zhou et al., 2021) . Measurable NAb titers against the early 184 pandemic virus (Victoria) were observed in the majority of participants tested 4 weeks after the first 185 dose (all infection-naïve), with a median 50% focus reduction neutralization titer (FRNT 50 ) of around 186 10 2 at this time point (Figure 2A) . For the variant viruses tested -beta (B.1.351), gamma (P.1) and 187 delta (B.1.167.2), there was very limited detection of NAb against beta and delta after one dose, but 188 only minimal reduction in titers against gamma compared with Victoria. Although the alpha variant 189 was not assayed in this set of experiments, extensive previous comparisons have indicated a 190 consistent drop in titer of around 3-fold (Supasa et al., 2021) . These titers declined up to 3-fold 191 following the peak, and were all markedly boosted following the second dose. Overall, a clear humoral response against the vaccine strain virus is induced by a single dose of vaccine across the cohort, although with NAbs at relatively low levels, especially against the beta 203 variant and delta variants. The peak NAb response following priming is followed by a decline during 204 the extended dose interval, most marked in naïve individuals, with a subsequent boost after the 205 second dose. Induction and maintenance of anti-spike T cell responses using the extended doing schedule 208 We next investigated the T cell response, using an established interferon-gamma (IFNγ) ELISpot 209 assay (Angyal et al., 2021; Ogbe et al., 2021) . The spike-specific T cell response was well maintained 210 during the 10 weeks following the primary vaccine, with no evidence of contraction, a phenomenon 211 that was equally seen in the previously infected and naïve groups ( Figure 2C ). T cells were boosted 212 by the second dose in both groups, with the greatest impact seen in the naïve group. In contrast, 213 responses to the non-spike T cell targets (M and N) showed only minimal changes over this period in 214 both groups ( Figure 2C ). Linear mixed-effects regression models were again performed to confirm 215 these results after adjusting for age and sex ( Table 2) . Overall, these data demonstrate the maintenance of cellular responses using the extended dosing 218 approach. We did not see a significant relationship between NAb and the T cell response to spike 219 (Figure S1 E-H). We next compared immune responses in the SARS-CoV-2 naïve cohort vaccinated using the longer 223 dosing interval with those vaccinated using the conventional 3-4 week (short) interval. We noted We also tested the impact of dosing interval on binding antibodies using spike, RBD and N as targets, 235 splitting the cohort by previous exposure. A clear advantage of the longer interval was seen again, 236 although only in the naïve cohort ( Figure 3C ). The previously infected cohort had equally high levels 237 of binding to spike and RBD regardless of vaccine regimen ( Figure 3C ). Generalised linear regression 238 models were performed to confirm these findings after adjustment for age, sex and previous 239 infection status, with separate models run for naïve and previously infected individuals ( the magnitude of antigen-specific B cell response 10 weeks after one dose in the extended interval 256 cohort was higher than 4 weeks after one dose in the short interval cohort, supporting continued B 257 cell development beyond 4 weeks after a prime. 4 weeks after the second dose, we saw a nearly 7-258 fold increase in the magnitude of the B cell response for the extended interval cohort compared to 259 the short interval, in parallel to the higher antibodies seen. Extension of the dosing interval did not lead to greater induction of T cell responses following the 262 second dose ( Figure 3E) We next compared the functionality of these T cell responses 4 weeks after the second dose in more 275 depth in 86 HCWs using intracellular cytokine staining (ICS). For this analysis, we selected only 276 participants with positive ELISpot assays, which we defined as >40/cells per million PBMC, the mean 277 of the DMSO negative values + 2 standard deviations, since in previous studies (Angyal et al., 2021; 278 Ogbe et al., 2021) we have observed that the smaller frequency populations are hard to detect by 279 flow cytometry, and also prone to inaccuracy due to low cell numbers. We saw a marked skewing in 280 the CD3 + T cell compartment towards a spike-specific CD4 + T cell response with the long dosing 281 interval, but more balance between CD4 + and CD8 + for the short interval ( Figure 4A , representative 282 gating strategy Figure S5 ). Further analysis showed that within the CD4 + compartment, spike-specific 283 responses for all cytokines tested were higher in naïve participants using the long dosing interval 284 compared to the short interval, whereas no differences were observed in the previously-infected 285 participants ( Figure 4B ). In contrast, within the CD8 + compartment there were no differences in 286 spike-specific responses except for CD8 + IFNγ responses which were lower in participants on the long 287 dosing interval, irrespective of pre-infection status ( Figure 4C ). These experiments revealed that infection-naïve recipients of the long dosing interval generated a 290 higher IL2 CD4 + response to spike compared with the short dosing interval, along with higher IFNγ 291 and TNF CD4 + responses ( Figure 4B ), whilst the CD8 + response was reversed with lower IFNγ 292 responses for the long interval ( Figure 4C ). By comparing the proportions of polyfunctional CD4 + T cells between the long and short dosing 295 groups, we observed an increased CD4 + T cell polyfunctionality in naïve participants who had 296 undergone the extended dosing interval compared to the short dosing interval ( Figure 4D) Throughout the analyses we noted differences in responses between naïve and previously infected 309 individuals. To explore this further, we used a larger dataset of all available unpaired data across the 310 time course (n=589 participants), and performed a side-by-side analysis of humoral and cellular 311 responses following first and second dose, including 13 weeks (3 months) after the second dose 312 where samples were available (Figure S3 A-D) . 4 weeks after a single dose of vaccination, previous 313 infection gives a significant advantage in magnitude of anti-S IgG antibody (approximately 8-14 fold, 314 Figure S3A ), and of T cell response (approximately 5-fold, Figure S3C ) (as reported in Angyal et al., 315 2021) . 4 weeks after 2 doses there is still a statistically significant difference in antibody and T cell 316 response between those with and without previous infection, although the magnitude is less 317 pronounced (2-3 fold). In those with previous infection, there was consistent evidence of higher 318 neutralisation activity against variants of concern using the RBD binding inhibition to ACE2 assay 319 ( Figure S3D ), suggesting "hybrid immunity" from natural infection plus vaccination gives the 320 strongest cross-reactive neutralisation. We have recently identified an association between the interval separating prior infection and first Future studies will evaluate the durability of the extended dosing interval. We observed maintenance of T cell responses over the same period ( Figure 5B ). These responses 344 showed a very modest, but statistically significant loss of activity against peptide pools from the 345 variant spike protein sequences, namely the beta (1.18-fold lower, P = 0.0001) and gamma (1.1-fold 346 lower, P = 0.0016) variants compared to wild-type sequence, but no loss of activity against delta 347 variant (fold change = 1.00, P = 0.2058) Figure 5C ). Overall these data recapitulate the data seen after a single dose. That is, there is a clear decline in The most significant observation of this study is that boosting after a longer interval leads to maintained immunogenicity. There is a distinct impact on the anti-spike responses, with an increase 406 in NAb, as seen for longer dosing intervals with the AstraZeneca 1222 vaccine, (Voysey et al., 2021) , 407 but a modest reduction in the IFNγ producing ELISpot response compared with conventional (short) 408 interval dosing. Although the difference in T cell response is quite small, it is reproducible with 409 different assays and was also seen in a study of an elderly UK cohort (Parry et al., 2021) . Both the 410 short and long dosing regimens result in induction of substantial T cell responses. The short dosing 411 interval gives a slightly higher IFNγ-producing T cell response, consistent with effector functions, 412 with both CD4 + and CD8 + contributions, whilst the longer dosing interval results in a CD4 + dominated 413 phenotype with marked IL2 production to spike. A recent preprint from Public Health England (PHE) 414 has confirmed higher antibody levels after extended dosing intervals compared to short, as well as 415 providing the first evidence of high vaccine effectiveness for extended dosing interval 416 (Amirthalingam et al., 2021) . As the authors discuss, some residual confounding between groups is 417 likely to remain, and further studies will follow. There is currently no agreed correlate of protection prior infection was much less evident than in naïve individuals. We also examined the influence of 458 age, sex and ethnicity in our cohort. Although these observations are limited by the numbers studied 459 and balance of the cohort, we did not observe any substantial effect in simple comparative or 460 multivariate analyses, except for a modest effect of older age associating with lower antibody levels 461 in naïve HCWs. It remains likely however that genetic effects such as HLA type do play a role as this 462 has been shown in other vaccine settings (Mentzer et al., 2015) , and the impact of exposure to other 463 beta coronaviruses remains to be further explored. In conclusion, the immunogenicity of longer regimens appears robust, and indeed for antibody 466 measurements, improved over the conventional 3-4 week regimen. We provide evidence that T cells 467 are induced and sustained during the longer period between doses in the 6-14 week regimen, but 468 there is an impact of dosing interval on the relative proportion of T cell subsets. Ongoing studies in 469 this cohort will monitor the durability of antibody and T cell responses 6 months after a 2 nd vaccine 470 dose delivered in an extended dosing interval, and response to 3 rd "booster" doses where given. For 471 policy makers, optimal dosing intervals may depend on community prevalence, population immunity 472 from natural infection, circulating variants of concern and vaccine supply. A short dosing interval 473 gives early protection, whereas an increased interval appears to improve peak neutralizing antibody 474 levels. Limitations  BNT162b2 vaccine with an extended interval between doses is highly protective  Antibody levels were higher after the extended regimen compared to the short regimen  The extended regimen enriches for virus-specific CD4+ T cells expressing IL-2  Antibody levels wane after each dose but B and T cell pools are maintained After giving a primary dose, delaying administration of a second dose of BNT162b2 COVID-19 vaccine up to 6-14 weeks continues to provide strong protection and contributes to favorable antibody, B cell, and T cell responses. J o u r n a l P r e -p r o o f Fold Change Higher serological responses and increased 1056 vaccine effectiveness demonstrate the value of extended vaccine schedules in combatting T-Cell and Antibody Responses to First BNT162b2 Vaccine Dose in Phase 1/2 trial of SARS-CoV-2 vaccine ChAdOx1 nCoV-1065 19 with a booster dose induces multifunctional antibody responses Humoral signatures of protective and pathological SARS-CoV-2 infection 1068 in children Sensitive and viable identification of antigen-specific CD8+ T cells by a flow cytometric assay 1071 for degranulation Antibody Responses after a Single Dose of SARS-CoV-1074 2 mRNA Vaccine Isolation and rapid sharing of the 2019 novel coronavirus (SARS-CoV-2) 1077 from the first patient diagnosed with COVID-19 in Australia Optimising T cell (re)boosting strategies for adenoviral and 1081 modified vaccinia Ankara vaccine regimens in humans Immunological memory to SARS-CoV-2 1084 assessed for up to 8 months after infection Antibody evasion by the P.1 strain of SARS-CoV-2 Department of Health and Social Care UK (2021) Optimising the COVID-19 vaccination programme 1089 for maximum short-term impact Evidence for antibody as a protective correlate for COVID-19 vaccines. medRxiv Antibody responses to the BNT162b2 mRNA vaccine in individuals 1097 previously infected with SARS-CoV-2 MHC class II invariant chain-adjuvanted viral vectored 1100 vaccines enhances T cell responses in humans COVID-19 vaccine coverage in health-care workers in England and 1112 effectiveness of BNT162b2 mRNA vaccine against infection (SIREN): a prospective, multicentre, 1113 cohort study IL-2 imprints human naive B cell fate towards plasma cell through ERK/ELK1-mediated BACH2 1116 repression Optimising the COVID-19 vaccination programme for maximum shortterm impact What level of neutralising antibody protects from COVID-1123 19? medRxiv Human Plasma Cell Generation: Coupling Differentiation and Proliferation by Enhancing MAPK-ERK 1126 Reduced neutralization of SARS-CoV-2 B.1.617 by vaccine and 1129 convalescent serum Antibody response to first BNT162b2 dose in 1132 previously SARS-CoV-2-infected individuals Correlates of protection against SARS-CoV-2 in rhesus macaques Searching for the human genetic 1137 factors standing in the way of universally effective vaccines. Philosophical transactions of the Royal Neutralizing antibodies elicited by the Ad26 Identification of resident memory CD8+ T cells with functional specificity 1145 for SARS-CoV-2 in unexposed oropharyngeal lymphoid tissue T cell assays differentiate clinical and subclinical SARS-CoV-2 infections from 1148 cross-reactive antiviral responses Differential immunogenicity of BNT162b2 or ChAdOx1 vaccines after extended-1151 interval homologous dual vaccination in older people Safety and Efficacy of the BNT162b2 mRNA Covid-19 Israel sees drop in Pfizer vaccine protection against infections Antigen-Specific Adaptive Immunity to SARS-CoV-2 in 1160 Acute COVID-19 and Associations with Age and Disease Severity Two doses of SARS-CoV-2 vaccination induce robust immune 1163 responses to emerging SARS-CoV-2 variants of concern Qualitative and quantitative analysis of adenovirus type 5 vector-1166 induced memory CD8 T cells: not as bad as their reputation Reduced neutralization of SARS-CoV-2 B.1.1.7 variant by 1169 convalescent and vaccine sera A human vaccine strategy based on chimpanzee adenoviral 1172 and MVA vectors that primes, boosts, and sustains functional HCV-specific T cell memory Early induction of functional SARS-CoV-2-specific T cells 1176 associates with rapid viral clearance and mild disease in COVID-19 patients Safety and Immunogenicity of a Heterologous Prime-Boost 1185 Ebola Virus Vaccine Regimen in Healthy Adults in the United Kingdom and Senegal. The Journal of 1186 infectious diseases Single-dose administration and the influence of 1189 the timing of the booster dose on immunogenicity and efficacy of ChAdOx1 nCoV-19 (AZD1222) 1190 vaccine: a pooled analysis of four randomised trials Evidence of escape of SARS-CoV-2 variant B.1.351 from 1194 natural and vaccine-induced sera