Antibody longevity and waning following COVID-19 vaccination in a 1-year longitudinal cohort in Bangladesh … – Nature.com

Study design, setting and study population

Participants of this cohort study were recruited from Sheikh Russel Gastroliver Institute & Hospital (SRGIH), a public sector health facility in Dhaka, Bangladesh when they visited the hospital to receive second dose of COVID-19 vaccines. Adults who received two doses of COVID-19 vaccines (OxfordAstraZeneca COVID19 vaccine or Covishield (viral vector-based vaccine), Pfizer-BioNTech (BNT162b2) (mRNA vaccine), Moderna (mRNA-1273) (mRNA vaccine) or Sinopharm (BBIBP-CorV) (inactivated whole virus vaccine)), were enrolled within 24weeks of receiving the second primary dose (visit 1) and followed-up at 4months (visit 2), 8months (visit 3) and 12months (visit 4). The inclusion criteria for participation were: (1) male or female adults (aged 18years and above), (2) able to understand and sign the informed consent form, (3) available and reachable by study staff for the entire period of the study; (4) agreeing to provide blood sample for the research study. The exclusion criteria were: (1) residence outside of Dhaka city; (2) suffering from long term severe illness such as cancer, chronic obstructive pulmonary disease (COPD) and chronic kidney disease (CKD).

Enrolment of participant started in September 2021 and ended in July 2022. Bangladesh received different types of COVID-19 vaccines at different time points either through COVAX or direct procurement, therefore availability of vaccinated individuals with any particular type of COVID-19 vaccine varied by time. Therefore, enrolment of vaccine groups varied by months. The Covishield was the first COVID-19 vaccine that was rolled out in Bangladesh and thus our first enrolled vaccine group was AstraZeneca vaccine. Administration of booster dose started in December 2021 in limited hospitals amongolder peoples (>60years of age) and age limit was gradually descended towards younger population. The participants of the AstraZaneca vaccine group received booster dose after almost 89months.

A structured questionnaire was utilized to collect data from each study participant at enrollment (visit 1, 24weeks after administration of 2nd primary dose) and each follow-up visit, i.e. at 4months (visit 2), 8months (visit 3) and 12months (visit 4) post second dose (Fig.1). Collected data included socio-demographic information (age, sex, education, migration background, ethnicity, marital status, household structure, occupation, and income), influenza-or COVID-like symptoms or presence of confirmed COVID-19 cases currently and in the past 6-months; co-morbidities (e.g. diabetes, hypertension, stroke, heart diseases and asthma). The vaccination type and administration dates were recorded. Information on receipt of booster dose was also collected during the follow-up phase. Height and weight were collected using stadiometer (Seca 217, Hamburg, Germany) and digital weighing scale (Camry-EB9063, China), respectively. Venous blood was collected at each visit in Lithium-heparin coated tubes (S-Monovette Plasma, Sarstedt AG & Co. KG, Nmbrecht, Germany), plasma was separated upon centrifugation, aliquoted, and stored at -80C until use.

The concentration of S-specific antibodies was determined in plasma by Elecsys Anti-SARS-CoV-2 S immunoassay kit (Roche DiagnosticsGmbH, Mannheim). This assay allowed quantitative determination of high affinity antibodies, predominantly IgG, but also IgA and IgM directed to the SARS-CoV-2 spike (S) protein receptor binding domain (RBD) in a double-antigen sandwich assay format on Cobas-e601 analyzer (Roche Diagnostics). According to the kit insert, the sensitivity of the Elecsys AntiSARSCoV2 S assay is 98.8%, clinical specificity is 99.98%, and the assay was found to have 92.3% positive agreement rate with a Vesicular Stomatitis Virus (VSV)-based pseudo-neutralization assay.

Nucleocapsid (N)-specific antibodies was determined in plasma by Elecsys Anti-SARS-CoV-2 immunoassay kit (Roche Diagnostics). The kit allows simultaneous detection of mature Nucleocapsid-specific IgM and IgG antibodies on an automated immunoassay analyzer (Cobas-e601, Roche Diagnostics). This is a qualitative assay that gives combined antibody titers of both IgM and IgG and does not differentiate between the two types. Based on the antibody cut-off index (COI), the serological response to SARS-CoV-2 is categorized as reactive (COI1.0, seropositive) and non-reactive (COI<1.0, seronegative). According to the kit insert, the Elecsys Anti-SARS-CoV-2 assay has 99.8% specificity and>99.5% sensitivity.

All participants vaccinated with viral vector-based vaccines (AstraZeneca) and mRNA vaccines (Pfizer and Moderna) were tested for N-antibodies to identify previous exposure to or infection with SARS-CoV-2. Participants immunized with Sinopharm vaccine (inactivated whole virus) were not tested for N antibodies as it is not possible to differentiate the N-antibodies induced by vaccination from those generated due to natural infection with SARS-CoV-2.

The study participants were defined as SARS-CoV-2 infected when tested positive by RT-PCR or showed COVID-like symptoms and concomitantly positive for N-antibodies. Individuals who did not exhibit COVID-like symptoms and were negative for N- antibodies or tested negative by RT-PCR were considered uninfected with SARS-CoV-2.

The authors affirm that all procedures involved in this research adhered to the ethical standards set by the pertinent national and institutional committees overseeing human experimentation, aligning with the Helsinki Declaration of 1975, revised in 2008. The study received approval from the institutional review board of icddr,b (PR-21069, dated 17 August 2021). Written informed consent was obtained from the participants.

The sample size was calculated based on the primary endpoint, i.e. to assess the persistence of SARS-CoV-2 S-specific antibody response following two primary doses of COVID-19 vaccines. A study by Shrotri M et al.12, demonstrated a decline of antibody titers by 19.7% from 0 to 21days to 2241days after administration of two doses of Pfizer vaccine. Based on this information, and considering statistical power of 80% and confidence interval of 95%, the estimated sample size was 64 per vaccine group. To reduce unknown bias due to different localities and socio demographic status, and unknown effects of COVID-19 vaccination, a design effect of 2 was added, which resulted in a sample size of 128. Considering 10% attrition rate, the final sample size was 140.8 (rounded to 141) per vaccine group and the total sample was 564. Despite rigorous efforts, we encountered difficulties in recruiting the target number of participants in the study. Within the duration of the study, administration of different COVID-19 vaccines was paused at different period. We were able to enroll only 452 participants, with the highest number in AstraZaneca vaccine group.

We showed basic demographic characteristics of the study participants, categorized by the respective vaccine types received (Pfizer, Moderna, AstraZeneca, and Sinopharm).

S-antibody data exhibited a right-skewed distribution in a histogram. To address this non-normal distribution, a natural log transformation was employed. To analyze the durability of S-antibody titers at different visits post-primary vaccination (primary endpoint), the participants across all vaccine groups were categorized into two groups: Group I consisted of individuals who remained uninfected with SARS-CoV-2 (defined above) and did not receive a 3rd dose of vaccine during the one-year study period. Group II included participants who got infected with SARS-CoV-2 (defined above), and did not receive a 3rd dose of vaccine during the study period. Furthermore, to assess the kinetics of S- antibody titers after the 3rd dose (secondary endpoint 1), the recipients were divided into 2 groups: Group III consisted of individuals who received the 3rd dose of vaccine between visit 2 and 3, and remained uninfected up to visit 4 (n=55); Group IV participants received the 3rd dose of vaccine between visits 2 and 3, but got infected with SARS-CoV-2 between visits 3 and 4 (n=12). Since the number of 3rd dose recipients was relatively small, the analysis in group III and IV was not stratified based on the type of COVID-19 vaccine received. Moreover, the AstraZeneca group was excluded from this analysis as this group received the 3rd dose between visit 3 and 4 (n=105), and could not be followed further after completion of 1-year follow-up to observe the durability of S-antibodies. To evaluate the mean differences in S-antibody titers between any two visits in each of the groups (groups I to IV), a multivariate regression model was employed. Repeated measure ANCOVA model was employed to examine the mean differences in S-antibody titers between multiple visits within a group. In order to determine the optimal model, age, sex, household income, occupation, body mass index (BMI), education, comorbidities, etc. were incorporated as covariates in the regression model. The effects of education and comorbidities was found minimal (<1%), thus, not included in the final regression model.

To assess the degree of protection from infection with SARS-CoV-2 (secondary endpoint 2) in mRNA and viral vector-based vaccine groups compared to inactivated vaccine as the reference group, Bayesian framework was adopted and Markov Chain Monte Carlo (MCMC) method was utilized to derive posterior distributions of model parameters. Furthermore, we applied model evaluation techniques and conducted sensitivity analyses to ensure the robustness of our results.

All statistical analyses were conducted using STATA (version 15) and Python (3.11), and GraphPad Prism was utilized for generating graphs. Significance was defined as a p-value of<0.05.

See original here:

Antibody longevity and waning following COVID-19 vaccination in a 1-year longitudinal cohort in Bangladesh ... - Nature.com