Effectiveness of the CoronaVac vaccine in the elderly ...

[Pages:43]medRxiv preprint doi: ; this version posted May 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-ND 4.0 International license .

1

Effectiveness of the CoronaVac vaccine in the elderly population during a P.1 variant-associated epidemic of COVID-19 in Brazil: A test-negative case-control study

Otavio T. Ranzani1,2*, Matt Hitchings3,4*, Murilo Dorion Nieto5, Tatiana Lang D'Agostini6, Regiane Cardoso de Paula6, Olivia Ferreira Pereira de Paula6, Edlaine Faria de Moura Villela6, Mario Sergio Scaramuzzini Torres7, Silvano Barbosa de Oliveira8,9, Wade Schulz10, Maria Almiron8, Rodrigo Said8, Roberto Dias de Oliveira11, Patricia Vieira da Silva12, Wildo Navegantes de Ara?jo,8,9,13, Jason R. Andrews14, Derek A.T. Cummings3,4, Albert I. Ko5,15, Julio Croda5,12,16

1-Barcelona Institute for Global Health, ISGlobal, Barcelona, Spain 2-Pulmonary Division, Heart Institute (InCor), Hospital das Clinicas HCFMUSP, Faculdade de Medicina, Universidade de S?o Paulo, S?o Paulo, SP, Brazil 3-Department of Biology, University of Florida, Gainesville, FL, USA 4-Emerging Pathogens Institute, University of Florida, Gainesville, FL, USA 5-Department of Epidemiology of Microbial Diseases, Yale School of Public Health, New Haven, CT, USA 6-Disease Control Coordination of the S?o Paulo State Department of Health, S?o Paulo, Brazil 7-Municipal Health Secretary of Manaus, Brazil, AM, Brazil 8-Pan American Health Organization, Bras?lia, DF, Brazil 9-Universidade de Bras?lia, Bras?lia, DF, Brazil 10-Department of Laboratory Medicine, Yale University School of Medicine, New Haven, CT, USA 11-State University of Mato Grosso do Sul - UEMS, Dourados, MS, Brazil 12-Universidade Federal de Mato Grosso do Sul - UFMS, Campo Grande, MS, Brazil 13-National Institute for Science and Technology for Health Technology Assessment, Porto Alegre, RS, Brazil 14-Division of Infectious Diseases and Geographic Medicine, Stanford University, Stanford, CA, USA 15-Instituto Gon?alo Moniz, Funda??o Oswaldo Cruz, Salvador, BA, Brazil 16-Fiocruz Mato Grosso do Sul, Funda??o Oswaldo Cruz, Campo Grande, MS, Brazil

*Authors contributed equally Authors contributed equally

Correspondence to: Prof Julio Croda, Universidade Federal de Mato Grosso do Sul and Funda??o Oswaldo Cruz, julio.croda@fiocruz.br

Keywords COVID-19; CoronaVac; inactivated whole-virus vaccine, P.1 variant; test-negative study; case-control study; Brazil

NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.

medRxiv preprint doi: ; this version posted May 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-ND 4.0 International license .

2

Abstract Background Mass vaccination is being used in response to coronavirus disease (COVID-19) epidemics, including those driven by emerging variants of concern. We evaluated the effectiveness of the inactivated whole-virus vaccine, CoronaVac, against symptomatic COVID-19 in the elderly population of S?o Paulo State, Brazil during widespread circulation of the P.1 variant. Methods We conducted a test-negative, case-control study of adults 70 years of age from S?o Paulo State from January 17 to April 29, 2021, during which vaccination with a two-dose regimen of CoronaVac was implemented. We identified RT-PCR-confirmed COVID-19 cases and controls who had a symptomatic illness with a negative RT-PCR test from national surveillance and state vaccination databases. Controls were pair-matched to cases by age category, sex, race, municipality, prior COVID-19 status, and date of RT-PCR testing. We estimated vaccine effectiveness, adjusted for age and comorbidities, using conditional logistic regression. Findings We selected 7,950 matched pairs with a mean age of 76 years from 26,433 COVID-19 cases and 17,622 test-negative controls. Adjusted vaccine effectiveness was 18.2% (95% CI, 0.0 to 33.2) and 41.6% (95% CI, 26.9 to 53.3) in the period 0-13 and 14 days, respectively, after the 2nd dose. Administration of a single vaccine dose was not associated with reduced odds of COVID-19. Vaccine effectiveness 14 days after the 2nd dose declined with increasing age and was 61.8% (95% CI 34.8 to 77.7), 48.9% (95% CI 23.3 to 66.0) and 28.0% (95% CI 0.6 to 47.9) among individuals 70-74, 75-79 and 80 years of age, respectively (pinteraction = 0.05). Interpretation CoronaVac was 42% effective in the real-world setting of extensive P.1 transmission, but significant protection was not observed until completion of the two-dose regimen. These findings underscore the need to maintain non-pharmaceutical interventions when mass vaccination with CoronaVac is used as part of an epidemic response. Funding Pan American Health Organization

medRxiv preprint doi: ; this version posted May 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-ND 4.0 International license .

3

Research in context

Evidence before this study We searched Pubmed for articles published from inception of the pandemic until May 10, 2021, with no language restrictions, using the search terms "vaccine", "COVID-19", "SARS-CoV-2", "elderly", "age", "older". There are no studies reporting the real-world effectiveness of COVID-19 vaccines against P.1, and few studies reporting real-world effectiveness of COVID-19 vaccines among elderly individuals.

Added value of this study In a population based matched test-negative case-control study conducted in a setting with widespread P.1 variant transmission, we observed that two-doses of CoronaVac was effective in preventing symptomatic COVID-19 among those aged 70 years or older, a population remarkably underrepresented in phase 3 clinical trials. There was no evidence for protection after the first dose. We observed a significant decline in effectiveness at older ages.

Implications of all the available evidence Our results showed that the inactivated COVID-19 vaccine CoronaVac likely needs the complete twodose schedule for protection against symptomatic disease among the elderly. Additionally, effectiveness of the two-dose schedule decreased with age. These results have direct implications for national vaccination programs using CoronaVac, highlighting that non-pharmaceutical interventions should be in place throughout vaccination campaigns. Further research is needed on how to optimize the vaccination of the very elderly (80 years), considering different vaccine types or dosing schedules by age.

medRxiv preprint doi: ; this version posted May 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-ND 4.0 International license .

4

Introduction The SARS-CoV-2 pandemic has caused 3.3 million deaths worldwide as of early May 2021,1 and elderly individuals have suffered disproportionate morbidity and mortality.2 Since the beginning of the pandemic the development of an effective vaccine against SARS-CoV-2 was a priority, and several vaccines have been tested and licensed for use. A key public health question is whether the licensed vaccines are effective in the elderly, who may have impaired immune responses,3 and who were underrepresented in the clinical trials.4,5

Brazil is one of the countries most affected by the COVID-19 pandemic, totaling more than 15 million cases and 421,000 deaths as of early May 2021.1,6 Variants of Concern (VOC) have likely played a role in the recent surges across Brazil.7 The P.1 VOC,8 which was first detected in Manaus and has accrued mutations associated with decreased neutralization,9,10 now consists of the majority of new infections in Brazil.7,11

Several vaccines against COVID-19 have been proven efficacious in Phase III trials and are being used in mass vaccination campaigns across the globe, including Sinovac's CoronaVac vaccine.4,12 Evidence for the effectiveness of vaccination in real-world setting has come primarily from countries with rapid roll-out of the vaccine, such as Israel13 and the United Kingdom.14 However, evidence of effectiveness in groups that were not well represented in clinical trials, such as the elderly, and against VOCs, remains limited.

We aimed to evaluate the real-world effectiveness of CoronaVac in the general population 70 years in S?o Paulo State, the most populous state in Brazil (~46 million inhabitants), where P.1 is the predominant circulating variant.

medRxiv preprint doi: ; this version posted May 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-ND 4.0 International license .

5

Methods Study setting The State of S?o Paulo (23?3S, 46?4'W) has 645 municipalities and 46 million inhabitants and its capital, S?o Paulo city, has 12 million inhabitants.15 Its population 70 years is projected as 3.23 million by 2020.15 S?o Paulo State reported 2,997,282 COVID-19 cases (cumulative incidence rate: 6,475 per 100,000 population) and 100,649 deaths (cumulative mortality: 217 per 100,000 population) by 9 May 2021.16 The State Secretary of Health of Sao Paulo (SES-SP) initiated its COVID-19 vaccination campaign on 17 January 2021 and is administering two vaccines during the study period, CoronaVac and ChAdOx1. As of 29 April 2021, 10.7 million doses (7.1 million first doses and 3.5 million second doses) have been administered in the State. We analysed only CoronaVac because of the low coverage of ChAdOx1 during the study period. The suggested interval for CoronaVac in Brazil during the study period was 21 or 28 days.

Study design We conducted a retrospective, test-negative,17,18 matched case-control study to estimate the effectiveness of CoronaVac in reducing the odds of the primary outcome of symptomatic RT-PCRconfirmed SARS-CoV-2 infections. The study population was individuals aged 70 years who had a residential address in S?o Paulo State and with complete information, which was consistent between data sources, on age, sex, and residence, and vaccination status and dates. For this study, we selected cases and matched controls who had a positive and negative SARS-CoV-2 RT-PCR test result, respectively, during the study period of January 17 to April 29, 2021.

The study design and statistical analysis plan were specified in advance of extracting information from data sources and are described in a publicly available protocol

medRxiv preprint doi: ; this version posted May 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-ND 4.0 International license .

6

() and the Supplement. The study was approved by the Ethical Committee for Research of Federal University of Mato Grosso do Sul (CAAE: 43289221.5.0000.0021).

Data Sources We identified eligible cases and controls, their demographic characteristics, and chronic comorbidities by linking the state laboratory testing registry of the public health laboratory network (GAL) and national healthcare and surveillance databases (e-SUS and SIVEP-Gripe). We determined the vaccination status by linking these databases with the state vaccination registries (Vacina J?). The record linkage was done by the S?o Paulo State Government ? PRODESP ? using CPF numbers (Brazilian citizens' unique identifier code) and sent to the researchers anonymized. Notification to these systems is compulsory in Brazil. We retrieved VOC information from the GISAID database.11

Selection of cases and matched controls Cases were selected from the study population who had symptomatic COVID-19, defined as a symptomatic individual with a positive SARS-CoV-2 RT-PCR test result from a respiratory sample that was collected during the study period and the absence of a positive RT-PCR test in the preceding 90-day period; and who did not receive a dose of ChAdOx1 vaccine before sample collection. Controls were selected from the study population who did not have a SARS-CoV-2 infection, defined as a symptomatic individual with a negative SARS-CoV-2 test result from a respiratory sample that was collected during the study period and the absence of a positive RT-PCR test in the subsequent 14-day period, the preceding 90-day period, or previously in the study period; and did not receive a dose of ChAdOx1 vaccine before sample collection. Symptomatic illness was defined as the presence of one or more

medRxiv preprint doi: ; this version posted May 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-ND 4.0 International license .

7

reported COVID-19 related symptoms19 with an onset within 0-10 days before the date of sample collection.

We matched one test-negative control to each case according to a time window of ?3 days between the case sample collection date; age category defined in 5-year age bands (70-74, 75-79, etc.); municipality of residence; race (defined as brown, black, yellow, white, or indigenous); and prior COVID-19-like illness (defined as having had at least one episode of symptomatic respiratory illness in surveillance databases from the beginning of the pandemic to the start of the study, regardless of viral test confirmation). Matching factors were chosen from variables that were associated with vaccination coverage or timing, and with SARS-CoV-2 infection risk or healthcare access (see protocol). Upon identification of each case, a single control was chosen from the set of all eligible matching controls, and each control was matched to at most one case.

Statistical analysis We analyzed the effectiveness of the full two-dose schedule of CoronaVac against symptomatic SARSCoV-2 infection, in the period of time starting 14 days after administration of the second dose. To understand the timing of vaccine effectiveness, we evaluated the association between vaccination with two doses in the period 0-13 days after administration of the second dose, and vaccination with a single dose in the period starting 14 days after administration of the first dose. Finally, we evaluated the exposure of receiving the first vaccine dose from 0 to 13 days before the sample collection date. An association during this period, where the vaccine likely has no or limited effectiveness,4,20,21 may serve as an indicator of unmeasured confounding in the effectiveness estimate. The reference group for vaccination status was individuals who had not received a first vaccine dose before the date of sample collection.

medRxiv preprint doi: ; this version posted May 21, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-ND 4.0 International license .

8

We used conditional logistic regression to estimate the odds ratio (OR) of vaccination among cases and controls. 1-OR provided an estimate of vaccine effectiveness under the assumptions of a test-negative design.22 We included as covariates in the adjusted model: age as a continuous variable, and COVID-19associated comorbidities (cardiovascular, renal, neurological, haematological, or hepatic comorbidities, diabetes, chronic respiratory disorder, obesity, or immunosuppression). We conducted one prespecified subgroup analysis by age groups, and five additional post hoc subgroup analyses, by sex, number of chronic comorbidities (none vs. at least one), the two most frequent chronic comorbidities (cardiovascular disease and diabetes), and region of residence ("Grande S?o Paulo" health region vs. others). We did not evaluate the pre-specified subgroup analysis by previous infection because of small numbers of previously infected individuals among case-control pairs. For these analyses, we assessed the difference in vaccine effectiveness 14 days after the second dose with an interaction term with each subgroup of interest. Finally, in an exploratory analysis, we evaluated vaccine effectiveness in weekly time windows following vaccination (0-6, 7-13, 14-20, 21-27, and 28 days following first and second dose).

Power calculation After generating matched case-control pairs and before performing the analyses, we simulated the power of the data set to identify a vaccine efficacy of 40% comparing those with two doses 14 days after the second dose to those who had not received a vaccine (see the protocol for details). After extracting the surveillance databases on 6 May 2021, we determined that the power of the study was 99.9%.

All analyses were done in R, version 4.0.2.

 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download