Evaluating the relationship between ciprofloxacin prescription and non-susceptibility in Salmonella Typhi in Blantyre, Malawi: an observational study

BACKGROUND

Ciprofloxacin is the first-line drug for treating typhoid fever in many countries in Africa with a high disease burden, but the emergence of non-susceptibility poses a challenge to public health programmes. Through enhanced surveillance as part of vaccine evaluation, we investigated the occurrence and potential determinants of ciprofloxacin non-susceptibility in Blantyre, Malawi.

METHODS

We conducted systematic surveillance of typhoid fever cases and antibiotic prescription in two health centres in Blantyre, Malawi, between Oct 1, 2016, and Oct 31, 2019, as part of the STRATAA and TyVAC studies. In addition, blood cultures were taken from eligible patients presenting at Queen Elizabeth Central Hospital, Blantyre, as part of routine diagnosis. Inclusion criteria were measured or reported fever, or clinical suspicion of sepsis. Microbiologically, we identified Salmonella enterica serotype Typhi (S Typhi) isolates with a ciprofloxacin non-susceptible phenotype from blood cultures, and used whole-genome sequencing to identify drug-resistance mutations and phylogenetic relationships. We constructed generalised linear regression models to investigate associations between the number of ciprofloxacin prescriptions given per month to study participants and the proportion of S Typhi isolates with quinolone resistance-determining region (QRDR) mutations in the following month.

FINDINGS

From 46 989 blood cultures from Queen Elizabeth Central Hospital, 502 S Typhi isolates were obtained, 30 (6%) of which had either decreased ciprofloxacin susceptibility, or ciprofloxacin resistance. From 11 295 blood cultures from STRATAA and TyVAC studies, 241 microbiologically confirmed cases of typhoid fever were identified, and 198 isolates from 195 participants sequenced (mean age 12·8 years [SD 10·2], 53% female, 47% male). Between Oct 1, 2016, and Aug 31, 2019, of 177 typhoid fever cases confirmed by whole-genome sequencing, four (2%) were caused by S Typhi with QRDR mutations, compared with six (33%) of 18 cases between Sept 1 and Oct 31, 2019. This increase was associated with a preceding spike in ciprofloxacin prescriptions. Every additional prescription of ciprofloxacin given to study participants in the preceding month was associated with a 4·2% increase (95% CI 1·8-7·0) in the relative risk of isolating S Typhi with a QRDR mutation (p=0·0008). Phylogenetic analysis showed that S Typhi isolates with QRDR mutations from September and October, 2019, belonged to two distinct subclades encoding two different QRDR mutations, and were closely related (4-10 single-nucleotide polymorphisms) to susceptible S Typhi endemic to Blantyre.

INTERPRETATION

We postulate a causal relationship between increased ciprofloxacin prescriptions and an increase in fluoroquinolone non-susceptibility in S Typhi. Decreasing ciprofloxacin use by improving typhoid diagnostics, and reducing typhoid fever cases through the use of an efficacious vaccine, could help to limit the emergence of resistance.

FUNDING

Wellcome Trust, Bill & Melinda Gates Foundation, and National Institute for Health and Care Research (UK).

Characterization of Enzyme-Linked Immunosorbent Assay (ELISA) for Quantification of Antibodies against Salmonella Typhimurium and Salmonella Enteritidis O-Antigens in Human Sera

Nontyphoidal Salmonella (NTS) is a leading cause of morbidity and mortality caused by enteric pathogens worldwide in both children and adults, and vaccines are not yet available. The measurement of antigen-specific antibodies in the sera of vaccinated or convalescent individuals is crucial to understand the incidence of disease and the immunogenicity of vaccine candidates. A solid and standardized assay used to determine the level of specific anti-antigens IgG is therefore of paramount importance. In this work, we presented the characterization of a customized enzyme-linked immunosorbent assay (ELISA) with continuous readouts and a standardized definition of EU/mL. We assessed various performance parameters: standard curve accuracy, dilutional linearity, intermediate precision, specificity, limits of blanks, and quantification. The simplicity of the assay, its high sensitivity and specificity coupled with its low cost and the use of basic consumables and instruments without the need of high automation makes it suitable for transfer and application to different laboratories, including resource-limiting settings where the disease is endemic. This ELISA is, therefore, fit for purpose to be used for quantification of antibodies against Salmonella Typhimurium and Salmonella Enteritidis O-antigens in human samples, both for vaccine clinical trials and large sero-epidemiological studies.

Bacterial shedding and serologic responses following an outbreak of Salmonella Typhi in an endemic cohort

BACKGROUND

Salmonella enterica serovar Typhi (Salmonella Typhi) is the cause of typhoid fever. Salmonella Typhi may be transmitted through shedding in the stool, which can continue after recovery from acute illness. Shedding is detected by culturing stool, which is challenging to co-ordinate at scale. We hypothesised that sero-surveillance would direct us to those shedding Salmonella Typhi in stool following a typhoid outbreak.

METHODS

In 2016 a typhoid outbreak affected one in four residents of a Nursing School in Malosa, Malawi. The Department of Health asked for assistance to identify nursing students that might spread the outbreak to other health facilities. We measured IgG antibody titres against Vi capsular polysaccharide (anti-Vi IgG) and IgM / IgG antibodies against H:d flagellin (anti-H:d) three and six months after the outbreak. We selected participants in the highest and lowest deciles for anti-Vi IgG titre (measured at visit one) and obtained stool for Salmonella culture and PCR. All participants reported whether they had experienced fever persisting for three days or more during the outbreak (in keeping with the WHO definitions of 'suspected typhoid'). We tested for salmonellae in the Nursing School environment.

RESULTS

We obtained 320 paired serum samples from 407 residents. We cultured stool from 25 residents with high anti-Vi IgG titres and 24 residents with low titres. We did not recover Salmonella Typhi from stool; four stool samples yielded non-typhoidal salmonellae; one sample produced a positive PCR amplification for a Salmonella Typhi target. Median anti-Vi and anti-H:d IgG titres fell among participants who reported persistent fever. There was a smaller fall in anti-H:d IgG titres among participants who did not report persistent fever. Non-typhoidal salmonellae were identified in water sampled at source and from a kitchen tap.

CONCLUSION

High titres of anti-Vi IgG did not identify culture-confirmed shedding of Salmonella Typhi. There was a clear serologic signal of recent typhoid exposure in the cohort, represented by waning IgG antibody titres over time. The presence of non-typhoidal salmonellae in drinking water indicates sub-optimal sanitation. Developing methods to detect and treat shedding remains an important priority to complement typhoid conjugate vaccination in efforts to achieve typhoid elimination.

Understanding the epidemiology of iNTS disease in Africa in preparation for future iNTS- vaccine studies in endemic countries: Seroepidemiology in Africa of iNTS (SAiNTS) Study Protocol [Version 9.0]

Background: Non-typhoidal Salmonella (NTS) are a major cause of bloodstream infections amongst children in sub-Saharan Africa. A clear understanding of the seroepidemiology and correlates of protection for invasive NTS (iNTS) in relation to key risk factors (malaria, anaemia, malnutrition) in children in Africa is needed to inform strategies for disease control including vaccine implementation. Methodology: The SAiNTS study is a prospective community cohort study with paired serology samples from 2500 children 0-5 years at baseline and three months to measure age-stratified acquisition of lipopolysaccharide (LPS) O-antigen antibody (IgG) and serum bactericidal activity to the main serovars causing iNTS (Salmonella typhimurium and S. enteritidis). Children are selected from mapped and censused randomly selected households in Chikwawa, Malawi; an area with substantial malaria burden. The sampling framework is set within a malaria vaccination (RTS,S/ AS01) phase 4 cluster randomized trial (EPIMAL), allowing exploration of the impact of malaria vaccination on acquisition of immunity to NTS. Data on risk factors for invasive disease: malaria, anaemia and malnutrition as well as indicators of socioeconomic status and water and sanitation, will be collected using rapid diagnostic tests, anthropometry and electronic CRF’s. Stool sample analysis includes NTS culture and pan-Salmonella polymerase chain reaction to assess enteric exposure and biomarkers of environmental enteric dysfunction. Cases with iNTS disease will be followed up for comparison with community controls. Conclusions: The final cohort of 2500 children will allow investigation into the impact of risk factors for iNTS on the acquisition of immunity in children 0-5 years in an endemic setting, including comparisons to partner sero-epidemiology studies in three other sub-Saharan African sites. The data generated will be key to informing iNTS disease control measures including targeted risk factor interventions and vaccine implementation through investigation of correlates of protection and identifying windows of immune susceptibility in at-risk populations.

Waning of antibody levels induced by a 13-valent pneumococcal conjugate vaccine, using a 3 + 0 schedule, within the first year of life among children younger than 5 years in Blantyre, Malawi: an observational, population-level, serosurveillance study

BACKGROUND

Pneumococcal conjugate vaccines (PCVs) induce serotype-specific IgG antibodies, effectively reducing vaccine-serotype carriage and invasive pneumococcal disease (IPD). IgG production wanes approximately 1 month after vaccination in absence of serotype-specific exposure. With uncertainty surrrounding correlate of protection (CoP) estimates and with persistent vaccine-serotype carriage and vaccine-serotype IPD after PCV13 introduction, we aimed to profile population-level immunogenicity among children younger than 5 years in Blantyre, Malawi.

METHODS

For this serosurveillance study, we used a random subset of samples from a prospective population-based serosurvey in Blantyre, Malawi, done between Dec 16, 2016, and June 27, 2018. Sample selection was based on age category optimisation among children younger than 5 years, adequate sample volume, and available budget. We measured serotype-specific IgGs against the 13 vaccine serotypes (1, 3, 4, 5, 6A, 6B, 7F, 9V, 14, 18C, 19A, 19F, and 23F) and two non-vaccine serotypes (12F and 33F), as well as IgGs against three pneumococcal proteins (PsaA, NanA, and Ply), using ELISA and a direct-binding electrochemiluminescence-based multiplex assay. We estimated population-level, serotype-specific immunogenicity profiles using a linear spline regression model. Analyses included samples stratified to 20 3-month age strata (eg, age <3 months to 57-59 months).

FINDINGS

We evaluated 638 plasma samples: 556 primary samples and 82 unique secondary samples (each linked to one primary sample). Immunogenicity profiles revealed a consistent pattern among vaccine serotypes except serotype 3: a vaccine-induced IgG peak followed by waning to a nadir and subsequent increase in titre. For serotype 3, we observed no apparent vaccine-induced increase. Heterogeneity in parameters included age range at post-vaccination nadir (from 11·2 months [19A] to 27·3 months [7F]). The age at peak IgG titre ranged from 2·69 months (5) to 6·64 months (14). Titres dropped below CoPs against IPD among nine vaccine serotypes (1, 3, 4, 5, 6B, 7F, 9V, 18C, and 23F) and below CoPs against carriage for ten vaccine serotypes (1, 4, 5, 6B, 7F, 9V, 14, 18C, 19F, and 23F). Increasing antibody concentrations among older children and seroincident events were consistent with ongoing vaccine-serotype exposure.

INTERPRETATION

A 3 + 0 PCV13 schedule with high uptake has not led to sustained population-level antibody immunity beyond the first year of life. Indeed, post-vaccine antibody concentrations dropped below putative CoPs for several vaccine serotypes, potentially contributing to persistent vaccine-serotype carriage and residual vaccine-serotype IPD in Malawi and other similar settings. Policy decisions should consider alternative vaccine strategies, including a booster dose, to achieve sustained vaccine-induced antibody titres, and thus control.

FUNDING

Bill & Melinda Gates Foundation, Wellcome UK, and National Institute for Health and Care Research.

Quality of antibody responses by adults and young children to 13-valent pneumococcal conjugate vaccination and Streptococcus pneumoniae colonisation

Childhood pneumococcal conjugate vaccine (PCV) protects against invasive pneumococcal disease caused by vaccine-serotype (VT) Streptococcus pneumoniae by generating opsonophagocytic anti-capsular antibodies, but how vaccination protects against and reduces VT carriage is less well understood. Using serological samples from PCV-vaccinated Malawian individuals and a UK human challenge model, we explored whether antibody quality (IgG subclass, opsonophagocytic killing, and avidity) is associated with protection from carriage. Following experimental challenge of adults with S. pneumoniae serotype 6B, 3/21 PCV13-vaccinees were colonised with pneumococcus compared to 12/24 hepatitis A-vaccinated controls; PCV13-vaccination induced serotype-specific IgG, IgG1, and IgG2, and strong opsonophagocytic responses. However, there was no clear relationship between antibody quality and protection from carriage or carriage intensity after vaccination. Similarly, among PCV13-vaccinated Malawian infants there was no relationship between serotype-specific antibody titre or quality and carriage through exposure to circulating serotypes. Although opsonophagocytic responses were low in infants, antibody titre and avidity to circulating serotypes 19F and 6A were maintained or increased with age. These data suggest a complex relationship between antibody-mediated immunity and pneumococcal carriage, and that PCV13-driven antibody quality may mature with age and exposure.

Safety and immunogenicity of a typhoid conjugate vaccine among children aged 9 months to 12 years in Malawi: a nested substudy of a double-blind, randomised controlled trial

BACKGROUND

Typhoid fever is a substantial public health problem in Africa, yet there are few clinical trials of typhoid conjugate vaccine (TCV). We assessed immunogenicity and safety of Typbar TCV in Malawi.

METHODS

This substudy was nested within a phase 3, double-blind, parallel design, randomised controlled trial of TCV in children from Ndirande Health Centre in Ndirande township, Blantyre, Malawi. To be eligible, participants had to be aged between 9 months and 12 years with no known immunosuppression or chronic health conditions, including HIV or severe malnutrition; eligible participants were enrolled into three strata of approximately 200 children (9-11 months, 1-5 years, and 6-12 years), randomly assigned (1:1) to receive TCV or control (meningococcal serogroup A conjugate vaccine [MCV-A]) intramuscularly. Serum was collected before vaccination and at 28 days and 730-1035 days after vaccination to measure anti-Vi antibodies by ELISA. Because of COVID-19, day 730 visits were extended up to 1035 days. This nested substudy evaluated reactogenicity, safety, and immunogenicity by age stratum. Safety outcomes, analysed in the intention-to-treat population, included solicited adverse events within 7 days of vaccination (assessed on 3 separate days) and unsolicited adverse events within 28 days of vaccination. This trial is registered with ClinicalTrials.gov, NCT03299426.

FINDINGS

Between Feb 22 and Sept 6, 2018, 664 participants were screened, and 631 participants were enrolled and randomly assigned (320 to the TCV group and 311 to the MCV-A group). 305 participants in the TCV group and 297 participants in the MCV-A group were vaccinated. Among TCV recipients, anti-Vi IgG geometric mean titres increased more than 500 times from 4·2 ELISA units (EU)/mL (95% CI 4·0-4·4) at baseline to 2383·7 EU/mL (2087·2-2722·3) at day 28, then decreased to 48·0 EU/mL (39·9-57·8) at day 730-1035, remaining more than 11 times higher than baseline. Among MCV-A recipients, anti-Vi IgG titres remained unchanged: 4·3 EU/mL (4·0-4·5) at baseline, 4·4 EU/mL (4·0-4·7) on day 28, and 4·6 EU/mL (4·2-5·0) on day 730-1035. TCV and MCV-A recipients had similar solicited local (eight [3%] of 304, 95% CI 1·3-5·1 and three [1%] of 293, 0·4-3·0) and systemic (27 [9%] of 304, 6·2-12·6 and 27 [9%] of 293, 6·4-13·1) reactogenicity. Related unsolicited adverse events occurred similarly in TCV and MCV-A recipients in eight (3%) of 304 (1·3-5·1) and eight (3%) of 293 (1·4-5·3).

INTERPRETATION

This study provides evidence of TCV safety, tolerability, and immunogenicity up to 730-1035 days in Malawian children aged 9 months to 12 years.

FUNDING

Bill & Melinda Gates Foundation.

A clinical and molecular epidemiological survey of hepatitis C in Blantyre, Malawi, suggests a historic mechanism of transmission

Hepatitis C virus (HCV) is a leading cause of liver disease worldwide. There are no previous representative community HCV prevalence studies from Southern Africa, and limited genotypic data. Epidemiological data are required to inform an effective public health response. We conducted a household census-based random sampling serological survey, and a prospective hospital-based study of patients with cirrhosis and hepatocellular carcinoma (HCC) in Blantyre, Malawi. We tested participants with an HCV antigen/antibody ELISA (Monolisa, Bio-Rad), confirmed with PCR (GeneXpert, Cepheid) and used line immunoassay (Inno-LIA, Fujiribio) for RNA-negative participants. We did target-enrichment whole-genome HCV sequencing (NextSeq, Illumina). Among 96,386 censused individuals, we randomly selected 1661 people aged ≥16 years. Population-standardized HCV RNA prevalence was 0.2% (95% CI 0.1-0.5). Among 236 patients with cirrhosis and HCC, HCV RNA prevalence was 1.9% and 5.0%, respectively. Mapping showed that HCV RNA+ patients were from peri-urban areas surrounding Blantyre. Community and hospital HCV RNA+ participants were older than comparator HCV RNA-negative populations (median 53 vs 30 years for community, p = 0.01 and 68 vs 40 years for cirrhosis/HCC, p < 0.001). Endemic HCV genotypes (n = 10) were 4v (50%), 4r (30%) and 4w (10%). In this first census-based community serological study in Southern Africa, HCV was uncommon in the general population, was centred on peri-urban regions and was attributable for <5% of liver disease. HCV infection was observed only among older people, suggesting a historic mechanism of transmission. Genotype 4r, which has been associated with treatment failure with ledipasvir and daclatasvir, is endemic.

Hepatitis B vaccination impact and the unmet need for antiviral treatment in Blantyre, Malawi

BACKGROUND

Hepatitis B is the leading cause of cirrhosis and liver cancer in sub-Saharan Africa. To reduce hepatitis-associated mortality, antiviral treatment programmes are needed. We estimated prevalence, vaccine impact and need for antiviral treatment in Blantyre, Malawi to inform an effective public health response.

METHODS

We conducted a household study in Blantyre in 2016-2018. We selected individuals from a census using random sampling and estimated age-sex-standardised HBsAg seroprevalence. Impact of infant hepatitis B vaccination, which began in 2002, was estimated by binomial log-linear regression comparing individuals born before and after vaccine implementation. In HBsAg-positive adults, eligibility for antiviral therapy was assessed.

RESULTS

Of 97,386 censused individuals, 6,073 (median age 18 years; 56.7% female) were sampled. HBsAg seroprevalence was 5.1% (95% CI 4.3-6.1) among adults and 0.3% (0.1-0.6) among children born after vaccine introduction. Estimated vaccine impact was 95.8% (70.3-99.4). Of HBsAg-positive adults, 26% were HIV-positive. Among HIV-negative individuals, 3%, 6% and 9% were eligible for hepatitis B treatment by WHO, European and American hepatology association criteria, respectively.

CONCLUSIONS

Infant HBV vaccination has been highly effective in reducing HBsAg prevalence in urban Malawi. Up to 9% of HBsAg-positive HIV-negative adults are eligible, but have an unmet need, for antiviral therapy.

Safety and Efficacy of a Typhoid Conjugate Vaccine in Malawian Children

BACKGROUND

Typhoid fever caused by multidrug-resistant H58 Salmonella Typhi is an increasing public health threat in sub-Saharan Africa.

METHODS

We conducted a phase 3, double-blind trial in Blantyre, Malawi, to assess the efficacy of Vi polysaccharide typhoid conjugate vaccine (Vi-TCV). We randomly assigned children who were between 9 months and 12 years of age, in a 1:1 ratio, to receive a single dose of Vi-TCV or meningococcal capsular group A conjugate (MenA) vaccine. The primary outcome was typhoid fever confirmed by blood culture. We report vaccine efficacy and safety outcomes after 18 to 24 months of follow-up.

RESULTS

The intention-to-treat analysis included 28,130 children, of whom 14,069 were assigned to receive Vi-TCV and 14,061 were assigned to receive the MenA vaccine. Blood culture-confirmed typhoid fever occurred in 12 children in the Vi-TCV group (46.9 cases per 100,000 person-years) and in 62 children in the MenA group (243.2 cases per 100,000 person-years). Overall, the efficacy of Vi-TCV was 80.7% (95% confidence interval [CI], 64.2 to 89.6) in the intention-to-treat analysis and 83.7% (95% CI, 68.1 to 91.6) in the per-protocol analysis. In total, 130 serious adverse events occurred in the first 6 months after vaccination (52 in the Vi-TCV group and 78 in the MenA group), including 6 deaths (all in the MenA group). No serious adverse events were considered by the investigators to be related to vaccination.

CONCLUSIONS

Among Malawian children 9 months to 12 years of age, administration of Vi-TCV resulted in a lower incidence of blood culture-confirmed typhoid fever than the MenA vaccine. (Funded by the Bill and Melinda Gates Foundation; ClinicalTrials.gov number, NCT03299426.).

Cerebral malaria is associated with differential cytoadherence to brain endothelial cells

Sequestration of Plasmodium falciparum‐infected erythrocytes (IE) within the brain microvasculature is a hallmark of cerebral malaria (CM). Using a microchannel flow adhesion assay with TNF‐activated primary human microvascular endothelial cells, we demonstrate that IE isolated from Malawian paediatric CM cases showed increased binding to brain microvascular endothelial cells compared to IE from uncomplicated malaria (UM) cases. Further, UM isolates showed significantly greater adhesion to dermal than to brain microvascular endothelial cells. The major mediator of parasite adhesion is P. falciparum erythrocyte membrane protein 1, encoded by var genes. Higher levels of var gene transcripts predicted to bind host endothelial protein C receptor (EPCR) and ICAM‐1 were detected in CM isolates. These data provide further evidence for differential tissue binding in severe and uncomplicated malaria syndromes, and give additional support to the hypothesis that CM pathology is based on increased cytoadherence of IE in the brain microvasculature.

Cloning of an EF-P homologue from Bacteroides fragilis that increases B. fragilis glutamine synthetase activity in Escherichia coli

Investigations of possible regulators of Bacteroides fragilis glutamine synthetase (GS) activity were done in Escherichia coli using a compatible dual-plasmid system. The B. fragilis glnA gene, together with upstream and downstream flanking regions, was cloned onto the low copy number plasmid pACYC184 and expressed in the E. coli glnA ntrB ntrC deletion strain, YMC11. GS activity was monitored following co-transformation with a B. fragilis genomic library carried on the compatible plasmid pEcoR251. A gene was cloned that caused a twofold increase in B. fragilis GS activity but did not affect the activity of the E. coli GS enzyme or the B. fragilis sucrase (ScrL). Deletion of the B. fragilis glnA downstream region decreased basal levels of GS activity, but did not affect the ability of the cloned gene to increase the B. fragilis GS activity. Reporter gene analysis, using the B. fragilis glnA promoter region fused to the promoterless Clostridium acetobutylicum endoglucanase gene, showed no increase in reporter gene activity. This demonstrated that the increase in GS activity was not regulated at the transcriptional level, and that the cloned gene product was not affecting the copy number of the plasmid in trans. Sequence data indicated that the cloned gene had good amino acid identity to a range of elongation factor P (EF-P) proteins, the highest being to that of a Synechocystis sp (48%), and the least to Mycobacterium genitalium (27%). Amino acid identity to the E. coli EF-P was intermediate (37%). A possible role for EF-P in enhancing translation of the B. fragilis glnA mRNA is proposed.