key: cord-0734427-e9p9n2he authors: Thompson, M. G.; Burgess, J. L.; Naleway, A.; Tyner, H.; Yoon, S. K.; Meece, J.; Olsho, L. E. W.; Caban-Martinez, A.; Fowlkes, A. L.; Lutrick, K.; Groom, H. C.; Dunnigan, K.; Odean, M. J.; Hegmann, K.; Stefanski, E.; Edwards, L. J.; Schaefer-Solle, N.; Grant, L.; Ellingson, K.; Kuntz, J. L.; Zunie, T.; Thiese, M. S.; Ivacic, L.; Wesley, M. G.; Mayo Lamberte, J.; Sun, X.; Smith, M. E.; Phillips, A. L.; Groover, K. D.; Yoo, Y. M.; Gerald, J. K.; Brown, R. T.; Herring, M. K.; Joseph, G.; Beitel, S.; Morrill, T. C.; Mak, J.; Rivers, P.; Poe, B. P.; Lynch, B.; Tao Zhou, Y.; Zhang, J.; Kelleher, A. title: Prevention and Attenuation of COVID-19 by BNT162b2 and mRNA-1273 Vaccines date: 2021-06-03 journal: nan DOI: 10.1101/2021.06.01.21257987 sha: cd09bb2a605b78a368c023f3e9bd0b32920858c4 doc_id: 734427 cord_uid: e9p9n2he ABSTRACT BACKGROUND: Information is limited on messenger RNA (mRNA) BNT162b2 (Pfizer-BioNTech) and mRNA-1273 (Moderna) COVID-19 vaccine effectiveness (VE) in preventing SARS-CoV-2 infection or attenuating disease when administered in real-world conditions. METHODS: Prospective cohorts of 3,975 healthcare personnel, first responders, and other essential and frontline workers completed weekly SARS-CoV-2 testing during December 14 2020--April 10 2021. Self-collected mid-turbinate nasal swabs were tested by qualitative and quantitative reverse-transcription--polymerase-chain-reaction (RT-PCR). VE was calculated as 100%x (1-hazard ratio); adjusted VE was calculated using vaccination propensity weights and adjustments for site, occupation, and local virus circulation . RESULTS: SARS-CoV-2 was detected in 204 (5.1%) participants; 16 were partially ([≥]14 days post-dose-1 to 13 days after dose-2) or fully ([≥]14 days post-dose-2) vaccinated, and 156 were unvaccinated; 32 with indeterminate status (<14 days after dose-1) were excluded. Adjusted mRNA VE of full vaccination was 91% (95% confidence interval [CI]=76%--97%) against symptomatic or asymptomatic SARS-CoV-2 infection; VE of partial vaccination was 81% (95% CI=64%-90%). Among partially or fully vaccinated participants with SARS-CoV-2 infection, mean viral RNA load (Log10 copies/mL) was 40% lower (95% CI=16%-57%), the risk of self-reported febrile COVID-19 was 58% lower (Risk Ratio=0.42, 95% CI=0.18-0.98), and 2.3 fewer days (95% CI=0.8-3.7) were spent sick in bed compared to unvaccinated infected participants. CONCLUSIONS: Authorized mRNA vaccines were highly effective among working-age adults in preventing SARS-CoV-2 infections when administered in real-world conditions and attenuated viral RNA load, febrile symptoms, and illness duration among those with breakthrough infection despite vaccination. and illness duration. VE was estimated for full vaccination (≥14-days after mRNA vaccine dose-2) and partial vaccination (≥14-days post-dose-1 and up to 13-days post-dose-2). Days 1-13 following dose-1 were excluded because vaccination status was considered indeterminate. Hazard ratios were estimated by the Andersen-Gill extension of the Cox proportional hazards model, which accounted for time-varying vaccination status. Unadjusted VE was calculated as 100%×(1−hazard ratio). An adjusted VE model accounted for potential confounding in vaccination status using an inverse propensity of treatment weighting (IPTW) approach 5 . Generalized boosted regression trees were used to estimate individual propensities to be at least partially vaccinated during each study week conditional on baseline socio-demographic and health characteristics and the most recent reports of virus exposure and PPE use (Table 1 and Table_S2) 6 . Predicted propensities were then used to calculate stabilized IPTW weights. Cox models incorporated these stabilized weights as well as covariates for site and occupation and a daily indicator of local virus circulation (the percentage positive of all SARS-CoV-2 testing for local counties; Figure_S1). A sensitivity analysis removed person-days among participants with possible misclassification of vaccination or infection and at locations when local SARS-CoV-2 circulation fell below 3% as described in Supplementary_Appendix_Methods. Due to the relatively small number of breakthrough infections, for evaluation of possible vaccination attenuation effects, participants with RT-PCR-confirmed infections while partially or fully vaccinated were combined into a single vaccinated group and compared to participants who were unvaccinated when infected. The highest viral RNA load (Log 10 copies/mL) measured during RT-PCR-confirmed infections were compared using a Poisson model adjusted for days from symptom onset to specimen collection and for days in transit to the laboratory. Dichotomous outcomes were compared using binary log-logistic regression to calculate relative for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 3, 2021. ; risks RR). Comparisons of illness duration outcomes in days were made with Student's t-test assuming unequal variances. All analyses were conducted with SAS (version 9.4; SAS Institute) and R (version 4.0.2; R Foundation for Statistical Computing). After exclusion of 1,147 participants with laboratory documentation of SARS-CoV-2 infection before the start of the study period, the analytic sample consisted of 3,975 participants; CONSORT diagram is Figure_S2. Approximately one-half of the participants (50.8%) were from the three Arizona study sites (Table 1 ). The majority of participants were female (62.0%), aged 18-49 years (71.6%), White (86.3%), and non-Hispanic (82.8%) and had no chronic medical conditions (68.6%). Participants included primary healthcare providers (20.4%), such as physicians and other clinical leads , nurses, and other allied healthcare personnel (33.0%), first responders (20.6%), and other essential and frontline workers (26.1%). Over the 17-week study period, adherence to weekly surveillance reporting and specimen collection was high (median = 100%; interquartile range = 82%-100%). Most (3,179/3,975, 80.0%) of the participants received at least one dose of an authorized mRNA COVID-19 vaccine by April 10, 2021 (Table 1) This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 3, 2021. ; one vaccine dose was higher among participants in Minnesota or Oregon and participants who were female, aged ≥ 50 years, White, non-Hispanic, healthcare personnel, or had ≥ 1 chronic health condition. Mean hours of close contact with people suspected or confirmed to have COVID-19 was lower and the percentage of time using PPE was higher among vaccinated participants (Table 1) . Associations with additional covariates included in the vaccination propensity model are presented in Table_S2. Standardized mean differences between vaccinated and unvaccinated were well balanced after propensity weighting with a maximum difference of 0.09 (Figure_S3). SARS-CoV-2 infection was detected by RT-PCR in 204 (5.1%) participants, including 5 fully and 16 partially vaccinated, and 156 unvaccinated; 32 with indeterminate vaccination status (<14 days post-dose-1) were excluded. Of 93 genetically sequenced viruses, 10 were variants of concern (8 B.1.429, 1 B.1.1.7, 1 B.1.427) and 1 was a variant of interest (P.2) (Table_S3). Excluding 12 viruses among those with indeterminate vaccination, 10 viruses were detected among partially or fully vaccinated participants; of these, 3 (30.0%) were variants of concern (all B.1.429) compared to 7/70 (10.0%) among unvaccinated participants. RT-PCR-confirmed infection was higher among participants in Arizona, Florida, and Texas or and those who were males, Hispanics, or first responders (Table 1) ; however, infection outcome did not differ by reported hours of potential virus exposure and PPE use. The majority of RT-PCR-confirmed infections had COVID-19-like illness symptoms before or within 1 day of specimen collection (74.0%) or within 2-14 days after collection (13.2%); the remainder had other symptoms (2.0%) or were asymptomatic within 14 days before and after specimen for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Estimated adjusted VE of full vaccination against RT-PCR confirmed infection was 91% (95% confidence interval [CI] = 76%-97%); VE of partial vaccination against RT-PCR confirmed infection was 81% (95% CI=64%-90%) ( Table 2 ). Secondary VE estimates by mRNA vaccine product and age group are available in Table 2 . VE point estimates were unchanged in a sensitivity analysis that excluded periods of low local virus circulation (Table_S4). Characteristics of 16 participants who were partially or fully vaccinated at SARS-CoV-2 infection and 156 participants who were unvaccinated at the time of infection are listed in Table_S5. The percentage infected while partially or fully vaccinated was higher among participants in Arizona, Minnesota, and Utah and among healthcare personnel; otherwise, there were no other differences by socio-demographic, health, virus exposure, or PPE use (Table_S6). for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 3, 2021. ; Mean viral RNA load measured for 153 participants was not associated with participant characteristics, except for somewhat lower viral RNA load among first responders (Table_S6). The mean viral RNA load (Log 10 copies/mL) detected for SARS-CoV-2 infections was 3.8 among unvaccinated and 2.3 among partially or fully vaccinated participants; in an adjusted model, this represented 40.2% (95% CI=16.3%-57.3%) lower viral RNA load after at least partial vaccination (Table 3) . Examining viral RNA only among those vaccinated, mean viral RNA load detected declined after receipt of dose-1 (Figure_S4). The majority of infections among partially or fully vaccinated participants were detected for only a single week (75.0%), while the majority of infections among unvaccinated participants (72.4%) were detected for 2 or more consecutive weeks; this represents a 66% reduction in the relative risk of RT-PCR-detection for ≥ 2 consecutive weeks (Table 3) . Measures of COVID-19-like illness severity and duration were not associated with participant characteristics, with the exception of a lower mean of symptom duration at Texas and Utah and a lower percentage of febrile COVID-19-like illness at Florida and Utah sites (Table_S6). Among participants with RT-PCR-confirmed SARS-CoV-2 infections, only 25.0% of partially or fully vaccinated participants reported febrile COVID-19-like illness compared to 63.1% of unvaccinated participants; this represents a 58% reduction in the relative risk of febrile COVID-19-like illness following at least partial vaccination (Table 3) . Vaccinated participants also reported 6.4 fewer (95% CI =0.4-12.3) total days ill or symptomatic and 2.3 fewer (95% CI=0.8-3.7) days sick in bed with COVID-19-like illness than unvaccinated participants. for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 3, 2021. ; In prospective cohorts of 3,975 healthcare personnel, first responders, and other essential and frontline workers followed over 17 weeks in eight U.S. locations, mRNA COVID-19 vaccines (Pfizer-BioNTech's BNT162b2 and Moderna's mRNA-1273) were 91% (95% CI = 76%-97%) effective in preventing symptomatic and asymptomatic RT-PCR-confirmed SARS-CoV-2 infection; VE of partial vaccination was 81%. These estimates of VE in real-world conditions are consistent with findings from 1,2 and from a similar prospective study of healthcare personnel that also conducts routine SARS-CoV-2 testing 7 . Among small number of participants with breakthrough RT-PCR-confirmed SARS-CoV-2 infections despite vaccination, mRNA vaccines appeared to attenuate infection and disease in multiple ways. Specimens from participants who were partially or fully vaccinated at the time of infection had a 40% reduction in viral RNA detected and were less likely to be RT-PCR-positive for more than one week compared to infected, unvaccinated participants. The risk of infected individuals developing febrile COVID-19-like illness was 66% lower and illnesses were on average about 6 days shorter with 2 fewer days sick in bed among partially or fully vaccinated participants compared with unvaccinated participants. Reduced viral RNA presence following mRNA COVID-19 vaccination is consistent with another recent report 8 , and the combination of virologic and clinical effects we observed is consistent with previous findings of lower quantity and duration of viral RNA detection with milder COVID-19 9 . The mechanisms by which COVID-19 vaccination elicits disease attenuation are largely unknown but are likely due to recall of immunologic memory responses that reduce viral replication and accelerate elimination of virally infected 10 . The biologic plausibility of these benefits are supported by similar phenomena observed in other studies 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 . Our for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 3, 2021. findings are also consistent with reports of lower symptom severity among those with moderate COVID-19 following vaccination with the Ad26.COV2.S COVID-19 vaccine compared to those who received placebo in a randomized controlled trial 20 . Among this study's strengths are the focus on working-age adults without prior laboratorydocumented SARS-CoV-2 infections, use of weekly testing for SARS-CoV-2 infection and illness with high surveillance adherence, multi-method documentation of vaccination status, and estimation of VE using vaccination propensity weighting, continuous local virus circulation updates, potential virus exposure, and PPE use. The study's use of a standard synthetic RNA to conduct quantitative RT-PCR improves upon most prior studies that have relied upon cycle thresholds from real-time RT-PCR as a proxy for viral RNA load 9 . This study also has at least seven limitations. First, although our estimate of 81% VE for partial vaccination is similar to other reports 1, 2, 7, 21, 22 , this estimate is based on a relatively brief period of follow-up (median of 19 days partially vaccinated versus 69 days fully vaccinated per participant). Second, we may have overestimated VE if we disproportionately failed to detect infections among vaccinated participants due to vaccine attenuation of viral RNA load or reductions in RT-PCR sensitivity due to self-collection and shipping of specimens 23 . Third, the study has not completed genetic sequencing for all viruses. Fourth, due to the relatively small number of breakthrough infections observed, we could not differentiate attenuation effects associated with partial versus full vaccination. Similarly, sparse data reduced the precision of estimates, though the consistency of trends across measures affirms the direction of the overall effect. Fifth, due to sparse data and limited racial and ethnic diversity, we were also unable to fully examine or adjust for potential confounders of vaccine attenuation effects. Nonetheless, we stratified our participant recruitment to ensure a combination of participant characteristics by for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 3, 2021. ; occupation, age, and sex; we did not observe consistent associations between socio-demographic, health, virus exposure, or PPE use with either vaccination status, viral RNA load, or illness duration. Sixth, febrile symptoms and illness duration were measured using self-report, which can be subject to recall and confirmation biases. Yet, these findings are consistent with our virologic findings of reduced viral load and duration of detection. Finally, detection of viral RNA is not equivalent to isolation of infectious virus; however, lower RT-PCR cycle thresholds have been associated with the ability to isolate SARS-CoV-2 in culture 9 , and both viral quantity and duration of viral RNA detection are associated with infectivity and transmission in other viral infections 19,24-26 . If further data confirms that mRNA vaccination reduces the number of viral RNA particles and the duration of detection and this in turn blunts the infectivity of SARS-CoV-2, then mRNA vaccines are not only highly effective in preventing SARS-CoV-2 infection, but they may also mitigate the impact of breakthrough infections, which is especially important to essential and frontline workers given their potential to transmit the virus through frequent close contacts with patients, co-workers, and the public. Disclosure forms provided by the authors are available with the full text of this article at NEJM.org. for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 3, 2021. ; for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 3, 2021. ; . L e v i n e -T i e f e n b r u n M , Y e l i n I , K a t z R , e t a l . I n i t i a l r e p o r t o f d e c r e a s e d S A R S -C o V -2 v i r a l l o a d a f t e r i n o c u l a t i o n w i t h t h e B N T 1 6 2 b 2 v a c c i n e . N a t u r e M e d i c i n e 2 0 2 1 . D O I : 1 0 . 1 0 3 8 / s 4 1 5 9 1 -0 for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 3, 2021. ; for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 3, 2021. ; Abbreviations: Interquartile range (IQR), Healthcare personnel (HCP), First responders (FR), Essential and frontline workers (EFW), COVID-19-like illness (CLI); Messenger RNA (mRNA); Personal protective equipment (PPE) † Analytic sample excludes 1,147 participants with documented SARS-CoV-2 infection before enrollment or as part of surveillance. § Comparison of those who were vaccinated with at least one dose and those who were not, cohort locations for Portland, OR, Duluth, MN, Salt Lake City UT were combined compared to For 15 participants missing biological sex, it was imputed as the more common category (female). ¶ Occupation categories: Primary HCP (physicians, physician assistants, nurse practitioners, dentists), Other allied HCP (nurses, therapists, technicians, medical assistants, orderlies and all others providing clinical support in inpatient or outpatient settings), first responders (FR; firefighters, law enforcement, corrections, emergency medical technicians), essential and frontline workers (EFW; workers in hospitality, delivery, and retail; teachers; all other occupations that require contact within 3 feet of the public, customers, or co-workers as a routine part of their job). ** For 77 participants, who did not respond to the self-report question, they were imputed as none, pending further verification. † † Each month, participants were asked about close contacts and PPE use during the past 7 days. The mean of monthly responses during the study period were calculated. for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 3, 2021. ; https://doi.org/10.1101/2021.06.01.21257987 doi: medRxiv preprint ‡ ‡ Only applicable for participants indicating a potential exposure during the past 7 days. for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 3, 2021. for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 3, 2021. for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 3, 2021. Duration of Viral RNA Detection Abbreviations: Standard deviation (SD); Relative risk (RR) Confidence interval (CI) One specimen could not be tested due to insufficient volume ‡ Poisson model adjusted for days from symptom onset to specimen collection and days of shipping to laboratory § Febrile illness defined as self-reported fever, feverishness, chills, or measured temperature >38 degrees Celsius For illness indicators among unvaccinated participants, the denominator for febrile versus afebrile/asymptomatic excludes 7 participants with confirmed COVID-like illness (CLI) but who did not complete an illness survey to document symptoms (or 156-7=149); for illness duration, the same 7 participants do not have total illness duration due to the missing illness survey and an additional participant had an illness without resolution by 4/10/21 (or 149-1=148); for days sick in bed For illness indicators among vaccinated participants, 1 participant with confirmed CLI completed an illness survey but was missing an answer to days sick in bed * Contributing participants in vaccination categories do not equal the number of vaccination doses because participants must be in active surveillance and have met the vaccination criteria. Persondays during days 1-13 following dose-1 and the 32 associated RT-PCR-confirmed infections are excluded from this model because immunity was considered indeterminate. † Adjusted VE is inversely weighted for propensity to be vaccinated with doubly robust adjustment for local virus circulation, study location, and occupation (Supplementary_Appendix_Methods)for use under a CC0 license. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.