Background: Maternal immunisation to boost respiratory syncytial virus (RSV) antibodies in pregnant women, is a strategy being considered to enhance infant protection from severe RSV associated disease. However, little is known about the efficiency of transplacental transfer of RSV-specific antibodies in a setting with a high burden of malaria and HIV, to guide the implementation of such a vaccination program. Methods: Using a plaque reduction neutralization assay, we screened 400 pairs of cord and maternal serum specimens from pregnant women for RSV-specific antibodies. Participants were pregnant women of two surveillance cohorts: 200 participants from a hospital cohort in Kilifi, Coastal Kenya and 200 participants from a surveillance cohort in Siaya, Western Kenya. Transplacental transfer efficiency was determined by the cord to maternal titre ratio (CMTR). Logistic regression was used to determine independent predictors of impaired transplacental transfer of RSV-specific antibodies. Results: A total of 800 samples were screened from the 400 participants. At enrollment the median age was 25 years (Interquartile range (IQR): 21-31). Overall, transplacental transfer was efficient and did not differ between Kilifi and Siaya cohort (1.02 vs. 1.02; p=0.946) but was significantly reduced among HIV-infected mothers compared to HIV-uninfected mothers (mean CMTR: 0.98 vs 1.03; p=0.015). Prematurity <33 weeks gestation (Odds ratio [OR]: 0.23, 95% confidence interval [CI] 0.06-0.85; p=0.028), low birth weight <2.5 kgs (OR: 0.25, 95% CI: 0.07-0.94; p=0.041) and HIV infection (OR: 0.47, 95% CI:0.23-0.98; p=0.045) reduced efficiency of transplacental transfer among these women. Conclusions: Transplacental transfer of RSV-specific antibodies among pregnant women in Kenya is efficient. A consideration to integrate other preventive interventions with maternal RSV vaccination targeting infants born premature (<33 weeks gestation), with low birth weight <2.5 kgs, or HIV-infected mothers is likely to improve vaccine outcomes in this setting.
We utilized existing data and serum samples from two independent cohorts of pregnant women in Kilifi and Siaya County, Kenya. One cohort was a hospital-based surveillance investigating risk factors for severe morbidity and mortality in mothers and their infants in Kilifi, coastal Kenya, and the other, a cohort in Siaya County, Western Kenya, for surveillance of influenza disease among pregnant women and their infants. The surveillance to investigate risk factors for severe morbidity and mortality in mothers and their infants was set up by KEMRI-Wellcome Trust Research Programme (KWTRP) at the maternity ward of Kilifi County Hospital (KCH) and the Kilifi Health and Demographic Surveillance System (KHDSS) area in Coastal Kenya in 2011 24 . This surveillance was designed to observe 4600 births with approximately 2300 being residents of KHDSS. All mothers presenting for delivery at the maternity ward of KCH were invited to enroll. Routine clinical data were collected using a standardized questionnaire at admission to the maternity department and following delivery. During this surveillance, consent was sought from pregnant women presenting at KCH maternity ward to collect cord and maternal blood samples after delivery. These samples were securely stored at -80˚C at KEMRI-Wellcome Trust laboratories in Kilifi for molecular and serological testing for viral and bacterial pathogens. The KIPMAT surveillance enrolled 4047 pregnant women in 2018 and 2019 who had a cord and maternal blood samples collected at delivery and stored for future use. From the Western part of Kenya, in Siaya County, a cohort of pregnant women was set up through a collaboration between KEMRI-Centre for Global Health Research and the U.S. Centers for Disease Control and Prevention (CDC) Kenya in 2015. This surveillance included pregnant women recruited either from their homes or when they visited for antenatal care at Bondo sub-County or Siaya County Referral Hospital. Participants were enrolled at gestational age <20 weeks. These pregnant women were followed up weekly through a phone call or home visit to record any occurrence of influenza-like illness episodes. Blood samples were collected at enrolment and a maternal and cord blood at birth. If a pregnant woman was identified with cough or fever during follow up, a respiratory specimen was collected and screened for influenza virus type A and B and for RSV using molecular methods 25 . All participants were requested to deliver their children in the hospital where birth outcomes were recorded; thereafter, both the baby and the mother were followed up weekly for up to six months post-delivery to assess infection from respiratory viruses by testing nasal and throat swabs from symptomatic cases by RT-PCR. During the RSV RT-PCR procedure, nucleic acids were extracted from 100ul of the combination of nasopharyngeal and oropharyngeal specimens using the MagMAX™ Viral RNA Isolation Kit on the Kingfisher mL platform (Life Technologies, New York). A 5ul of the Nucleic acid extract was then tested for RSV in a 1-step real-time reverse-transcription polymerase chain reaction (rRT-PCR) assay, using the AgPath-ID One-step RT-PCR kit (Applied Biosystems, Foster City, California) using CDC’s primers and fluorescent-labelled hydrolysis probes. The assay was considered positive for RSV when exponential fluorescence curves crossed the assigned threshold at a cycle threshold value of <37.0. The influenza surveillance had initially proposed to recruit 2250 pregnant women in a period of 3–5 years from 2015, based on assumptions that, each participant was to be followed up for a period of six months during pregnancy and 15% of participants were expected to have at least one influenza-like illness (ILI) episode. The study changed to a surveillance in 2018 and continued to enrol participants beyond the proposed sample size. From 2018 to 2019, the influenza surveillance had recruited 1458 pregnant women from whom, 795 participants had a cord and maternal blood samples at delivery, and these were used as a sampling frame for this study. For Kilifi cohort, 200 women were randomly selected from the cohort registers based on the availability of meta-data and paired cord and maternal blood samples for births (including preterm births) that occurred in 2018 and 2019. For Siaya cohort, where we had additional meta data, we selected all women (n=106) with lab-confirmed RSV infection, HIV, malaria, or anaemia, and randomly selected the other 94 women with available blood (maternal and cord) sample. A total of 400 participants were selected (200 from each region). The sample size was estimated using Edgar C. Fieller methods of calculating confidence intervals for the ratio of two means. This sample size method used CMTRs of 1.03 (0.88–1.19) observed in women in rural Nepal 18 . Assuming that CMTR of RSV-specific antibodies among women in Kenya were similar to those of Nepalese women and both the cord and maternal antibody levels followed a Gaussian distribution, a sample size of 200 mother-infant pair was sufficient to detect a CMTR of 1.03 with a 95% confidence interval of 1.01–1.06. All blood samples were screened for RSV specific antibodies using an inhouse plaque reduction neutralization titre (PRNT) assay 26, 27 at KWTRP laboratories, Kilifi, Kenya. The PRNT procedure determines the concentration of functional antibodies from a human serum sample (or antibody preparation) required to induce 50% neutralization of a known titration of RSV virus using the Spearman Karber method 28 . The assay has two stages, the neutralization step and a detection step. In this assay, mother-cord pairs of blood samples were assayed in one plate without use of complement sera. In the neutralization step, each serum sample was repeatedly diluted 2-fold over ten consecutive dilutions and mixed with an equal volume of 50 plaque forming units (pfu) of RSV A2 virus. The virus-serum mixture (50μl per well) was dispensed over a confluent monolayer of Hep2 cells in a 96 well culture plate, incubated at 37°C for 48 hours. In the detection step, fixation of cells was done by addition of 100μl/well of fixation reagent (30% methanol+70% acetone). This was followed by addition of a primary antibody (RSV F protein mouse monoclonal-BIO-RAD, Catalogue# MCA490) solution diluted 1:500 in PBS with 2 hours incubation at 37°C, and an addition of a 100μl/well of an anti-mouse HRP-conjugated secondary antibody (170-5047 Immun-Star Goat Anti-Mouse (GAM)- IgG (H/L) polyclonal antibody HRP–BIORAD). Micro RSV plaques were stained brown by 100μl/well detection reagent consisting of 16 μl of hydrogen peroxide and 0.6ml of 3-amino-9-ethlycarbazole 3.3mg/ml solution and counted with an ELISPOT reader. Plaque counts generated by the ELISpot reader were copied and pasted onto an excel analysis template containing Spearman-Karber formula to generate the neutralizing antibody titre for each sample. The PRNT titre was calculated as a neutralizing dose (ND50) value as follows: log10ND50 = m – Δ (Σp – 0.5). Where m was the log10 dilution of the highest dilution of serum (i.e. log10 (1/10,240) = -4.01) Δ was the constant interval between dilutions expressed as log10 (i.e. log10 (2) = 0.3010). The reciprocal was considered as the final antibody titre. This assay procedure has been described elsewhere in detail 27 . An RSV group A human reference standard (RSV IS 16/284) 29 obtained from National Institute for Biological Standards and Control (NIBSC), Potters Bar, UK, and an inhouse pooled adult sera were incorporated into each assay run to check for antibody deterioration, standardization of sample titres and quality control. Written informed consent to collect samples and data for storage and use in other studies was obtained from all participants through the parent studies, i.e., the influenza cohort surveillance (SERU #2880; CDC IRB number 6709) and the surveillance for risk factors cohort (SERU #1778). Ethical approval to screen samples for RSV-specific antibodies and use of data from the parent studies for this study was granted by the KEMRI Scientific and Ethical Review Unit Committee (SERU #3716). All methods were carried out in accordance with relevant guidelines and regulations. Separate analysis was done for each cohort and with combined data from both cohorts. All PRNT titres were log normalised (log base2) before analysis. The efficiency of transplacental transfer of RSV-specific antibodies was calculated for each mother-infant pair of blood samples. A CMTR (i.e., PRNT titre cord/PRNT titre maternal blood) of ≥1 was considered normal or efficient, CMTR <1 but ≥0.8 as moderately impaired and <0.8 as severely impaired or poor. Duration of transplacental transfer was calculated as gestational age at delivery minus 28; where 28 th week was estimated as the gestational age when a transplacental transfer of IgG antibodies begins during pregnancy. For this analysis, preterm birth (PTB) was defined as baby born alive before 37 weeks of pregnancy are completed and very early PTB as baby born <33 weeks of gestation. The difference in CMTR, cord or maternal RSV PRNT titres between HIV-infected versus HIV-uninfected mothers and RSV-infected infants vs. uninfected infants were analysed using a two-sample paired t-test. The Chi-square test was used to compare characteristics of women between Kilifi and Siaya cohort; and was also applied to determine the association between maternal/infant characteristics (HIV infection, malaria infection, RSV infection, anaemia, education level, occupation, gestational age at delivery and birth weight) and efficiency of transplacental transfer of RSV-specific antibodies. Logistic regression adjusted for each variable category (HIV infection, malaria infection, gestational age at delivery, gravida, birthweight and RSV infection during pregnancy) was used to determine independent predictors of an impaired transplacental transfer of RSV specific antibodies. In the multivariable logistic regression model, transplacental transfer efficiency was used as a binary outcome (normal [CTMR ≥1] vs impaired [CMTR <1] transfer). All data analysis was conducted using STATA version 15.0 (Stata Corp, College Station, Texas). However, a R Statistical software version 4.1.1 which is on open-access, can perform the equivalent analysis. To replicate the same analysis in R, we advise the user to import the CSV version of the data and follow the steps provided in the STATA do-file.
N/A