Pieter Bos

Anticipating rotavirus vaccines: epidemiology and surveillance of rotavirus in South Africa

Rotavirus infection is associated with acute infantile gastroenteritis in infants and young children globally. In South Africa, rotavirus infection has been shown to be associated with approximately one-quarter of all diarrhoeal admissions to hospital. Rotavirus infection predominantly occurs in infants less than 12 months of age (75%) and has a peak of shedding during the cooler, drier months of the year. A secondary peak during the spring has been observed. Multiple infections with rotavirus and at least one other microbial agent are common. The circulating VP7 serotypes and VP4 genotypes have been determined in various regions of South Africa and show a geographic specific distribution. A decade previously, P[8]G1 or G4 strains predominated, and P[4]G2 strains occurred in an epidemic pattern in one region. More recently, rotavirus strains with P[6] genotype have become common and novel VP7/VP4 genotype combinations are occurring across the country. G9 strains have been reported from Cape Town to Vendaland. The circulating rotavirus types observed in this study add to the knowledge of the natural history of rotavirus infection and provide the groundwork to consider future vaccine strategies.

Co-administration study in South African infants of a live-attenuated oral human rotavirus vaccine (RIX4414) and poliovirus vaccines.

A double-blind, placebo-controlled phase II trial (e-Track 444563-014/NCT00346892) was conducted in South Africa to evaluate the co-administration of RIX4414 (live-attenuated human G1P[8] rotavirus vaccine) and oral poliovirus vaccine (OPV) administered simultaneously. Healthy infants (n=450) were randomized into three groups (RIX4414+OPV, RIX4414+IPV or Placebo+OPV) to receive two oral doses of RIX4414/placebo with OPV or IPV using two vaccination schedules (6-10 weeks and 10-14 weeks). Serum anti-rotavirus IgA antibodies (ELISA) and neutralizing antibodies (micro-neutralization assay) to poliovirus serotypes 1, 2 and 3 were measured. Co-administration of RIX4414 with OPV did not result in a decrease in the high sero-protection rates against poliovirus serotypes 1, 2 and 3 detected after the third OPV dose (98-100%). The anti-rotavirus IgA antibody sero-conversion rates were higher for the 10-14 weeks schedule (55-61%) compared to the 6-10 weeks schedule (36-43%). Solicited symptoms were reported at similar rates between RIX4414 and placebo groups and no serious adverse events related to RIX4414 were reported. This study provided evidence that RIX4414 can be co-administered with routine EPI immunizations including OPV and that two doses of RIX4414 were well tolerated and immunogenic in South African infants.

Human rotavirus vaccine Rotarix™ provides protection against diverse circulating rotavirus strains in African infants: a randomized controlled trial

BackgroundRotaviruses are the most important cause of severe acute gastroenteritis worldwide in children <5 years of age. The human, G1P[8] rotavirus vaccine Rotarix™ significantly reduced severe rotavirus gastroenteritis episodes in a Phase III clinical trial conducted in infants in South Africa and Malawi. This paper examines rotavirus vaccine efficacy in preventing severe rotavirus gastroenteritis, during infancy, caused by the various G and P rotavirus types encountered during the first rotavirus-season.MethodsHealthy infants aged 5–10 weeks were enrolled and randomized into three groups to receive either two (10 and 14 weeks) or three doses of Rotarix™ (together forming the pooled Rotarix™ group) or three doses of placebo at a 6,10,14-week schedule. Weekly home visits were conducted to identify gastroenteritis episodes. Rotaviruses were detected by ELISA and genotyped by RT-PCR and nucleotide sequencing. The percentage of infants with severe rotavirus gastroenteritis caused by the circulating G and P types from 2 weeks post-last dose until one year of age and the corresponding vaccine efficacy was calculated with 95% CI.ResultsOverall, 4939 infants were vaccinated and 4417 (pooled Rotarix™ = 2974; placebo = 1443) were included in the per protocol efficacy cohort. G1 wild-type was detected in 23 (1.6%) severe rotavirus gastroenteritis episodes from the placebo group. This was followed in order of detection by G12 (15 [1%] in placebo) and G8 types (15 [1%] in placebo). Vaccine efficacy against G1 wild-type, G12 and G8 types were 64.1% (95% CI: 29.9%; 82%), 51.5% (95% CI:-6.5%; 77.9%) and 64.4% (95% CI: 17.1%; 85.2%), respectively. Genotype P[8] was the predominant circulating P type and was detected in 38 (2.6%) severe rotavirus gastroenteritis cases in placebo group. The remaining circulating P types comprised of P[4] (20 [1.4%] in placebo) and P[6] (13 [0.9%] in placebo). Vaccine efficacy against P[8] was 59.1% (95% CI: 32.8%; 75.3%), P[4] was 70.9% (95% CI: 37.5%; 87.0%) and P[6] was 55.2% (95% CI: -6.5%; 81.3%)ConclusionsRotarix™ vaccine demonstrated efficacy against severe gastroenteritis caused by diverse circulating rotavirus types. These data add to a growing body of evidence supporting heterotypic protection provided by Rotarix™.Trial registration numberNCT00241644

Efficacy and immunogenicity of two or three dose rotavirus-vaccine regimen in South African children over two consecutive rotavirus-seasons: A randomized, double-blind, placebo-controlled trial

BACKGROUND Human rotavirus vaccine (HRV; i.e., Rotarix) reduced the incidence of severe rotavirus gastroenteritis (RVGE) by 77% (95% Confidence interval: 56-88%) during the first year of life in South Africa. Persistence of HRV-derived protection against RVGE during subsequent rotavirus seasons, although evident in industrialized settings, remains to be established in African settings. This study reports on the efficacy of HRV against severe RVGE over two consecutive rotavirus seasons in South African children. METHODS A prospective, double-blind, placebo controlled multi-centered trial in South Africa and Malawi randomly assigned infants in a 1:1:1 ratio to receive either two (10 and 14 weeks; HRV_2D) or three (6, 10 and 14 weeks; HRV_3D) doses of HRV or placebo. The primary analysis involved pooling of HRV_2D and HRV_3D arms. Episodes of gastroenteritis caused by wild-type rotavirus were identified through active follow-up surveillance and graded by the Vesikari scale. RESULTS 1339 infants (447 in the HRV_2D group, 447 in the HRV_3D group and 445 in the placebo group) were enrolled in Year 2 of the study, including 1035 (77.3%) who were followed up over two consecutive rotavirus seasons (i.e., Cohort 2 subjects). Rotarix was associated with ongoing protection against severe RVGE, preventing 2.5 episodes per 100 vaccinated children over two consecutive rotavirus seasons; vaccine efficacy: 59% (95% Confidence interval: 1-83%). An exploratory analysis indicated better immunogenicity (among Cohort 1 subjects) and a higher point-efficacy estimate over two seasons in the HRV_3D compared to HRV_2D arms of the study in Cohort 2 subjects. CONCLUSION Rotarix is associated with significant reductions in severe gastroenteritis episodes through 2 years of life among South African children. Further research is needed to determine the optimal dosing schedule of Rotarix in providing long-term protection against rotavirus illness in African children.

Safety, Reactogenicity, and Immunogenicity of Human Rotavirus Vaccine RIX4414 in Human Immunodeficiency Virus-positive Infants in South Africa.

Background: Rotavirus and human immunodeficiency virus (HIV) infections are a cause of great public health concern in developing countries. The current study evaluated the safety, reactogenicity, and immunogenicity of RIX4414 vaccine in asymptomatic or mildly symptomatic (clinical stages I and II according to WHO classification) HIV-infected South African infants. Methods: A total of 100 HIV-positive infants aged 6 to 10 weeks enrolled in this double-blind, 1:1 randomized, placebo-controlled study were allocated into 2 groups to receive 3 doses of RIX4414 vaccine/placebo according to a 0-, 1-, and 2-month schedule. Routine vaccines were concomitantly administered. Solicited and unsolicited symptoms were recorded for 15 and 31 days after each dose, respectively. Serious adverse events were recorded throughout the study period. Serum antirotavirus IgA concentrations (enzyme-linked immunosorbent assay, cut-off ≥20 U/mL) and the immunodeficiency status were determined at screening and 2 months post-Dose 3. Stool samples were analyzed for rotavirus using enzyme-linked immunosorbent assay at predetermined points and during diarrhea episodes. Results: All symptoms (solicited and unsolicited) occurred at a similar frequency in both groups. Six fatal serious adverse events in RIX4414 and 9 in placebo groups were reported. At 2 months post-Dose 3, the seroconversion rates were 57.1% (95% CI: 34–78.2) in RIX4414 and 18.2% (95% CI: 5.2–40.3) in the placebo group. The mean absolute CD4+ cell count, CD4+ percentage, and HIV-1 viral load were comparable in both groups at screening and 2 months post-Dose 3. Rotavirus shedding peaked at Day 7 after Dose 1 of RIX4414 with prolonged shedding was observed in 1 infant only. Conclusions: Three doses of RIX4414 vaccine was tolerated well by the South African HIV-positive infants. A satisfactory immune response was mounted without aggravating their immunologic or HIV condition.

Seroepidemiological study of genogroup I and II calicivirus infections in South and southern Africa.

Diarrhoea is associated with the daily death of between 180 and 200 children under the age of 5 years in South Africa. Until recently, many cases and outbreaks of diarrhoea were not associated with a known aetiologic agent. Previous studies using baculovirus‐expressed Norwalk virus (NV) and Mexico virus (MxV) capsid antigens have shown that human calicivirus infection is common in South Africa. In this study, our surveillance was extended to different populations, as well as to four other southern African countries: Namibia, Angola, Zimbabwe, and Mozambique. More than 1,700 specimens, some involved in previous cohort studies of infectious diseases, were enrolled in the surveillance. The overall seroprevalence of antibody against NV was >90% for all cohorts except for Mozambican refugees that had 83.8% sero‐positivity. The MxV antibody prevalence was higher than NV, with >95% positivity for all cohorts, except for one in Namibia that had 81% exposure. This study is one of only a few reporting on the concurrent incidence of NV and MxV infections in a cohort study, and has determined that small round structured viruses are prevalent in the local populations of South and Southern Africa. These agents may account for a number of previously unknown or unidentified causes of diarrhoeal illness, in both adults and children, in southern Africa. J. Med. Virol. 59:227–231, 1999.

Comparison of 2 Different Regimens for Reactogenicity, Safety, and Immunogenicity of the Live Attenuated Oral Rotavirus Vaccine RIX4414 Coadministered with Oral Polio Vaccine in South African Infants

BACKGROUND A phase II, randomized, double-blind, placebo-controlled study was conducted in South Africa during 2003-2004 to evaluate the safety, reactogenicity, and immunogenicity of 2 regimens of the live attenuated oral human rotavirus vaccine RIX4414 when coadministered with the Expanded Program on Immunization childhood vaccines, including oral polio vaccine. METHODS Healthy infants were randomized (2:2:1) to receive either 2 doses of RIX4414 (n = 190; at 10 and 14 weeks, with placebo at 6 weeks), 3 doses of RIX4414 (n = 189; at 6, 10, and 14 weeks), or 3 doses of placebo (n = 96), all with concomitant routine vaccinations. The antirotavirus IgA seroconversion rate was assessed using enzyme-linked immunosorbent assay at 2 months after the last dose of RIX4414 or placebo. Antipolio types 1, 2, and 3 antibodies were measured using a virus neutralization assay. Solicited symptoms were recorded for 15 days after each dose. RESULTS The antirotavirus IgA seroconversion rates were similar in the RIX4414 2- and 3-dose groups (44.3% and 44.4%, respectively; P = .544, by 1-sided Fisher exact test) and antirotavirus IgA geometric mean concentrations were also comparable. Seroprotection rates for antipolio types 1, 2, and 3 antibodies were high (93%-100%) and were not significantly different among groups. Solicited symptoms reported within 15 days after vaccination were similar in all groups. CONCLUSIONS The immune seroconversion response to the RIX4414 vaccine with 3 doses was not superior to the 2-dose regimen. There was no interference by either regimen with antibody response to oral polio vaccine, and RIX4414 was well tolerated when given with routine vaccinations.

Characterization and Molecular Epidemiology of Rotavirus Strains Recovered in Northern Pretoria, South Africa during 2003–2006

Rotavirus infection is the most common cause of severe dehydrating gastroenteritis in infants and young children and remains a significant clinical problem worldwide. The severity and the burden of rotavirus disease could be reduced through the implementation of an effective vaccine. The aim of this study was to characterize rotavirus strains circulating in the local community as part of an ongoing hospital burden of disease study when a G1P[8] rotavirus vaccine candidate was being evaluated in the same community. From 2003 through 2006, 729 rotavirus-positive stool specimens were collected from children <5 years of age who were treated for diarrhea at Dr George Mukhari Hospital, Ga-Rankuwa, South Africa. Molecular characterization of the strains was performed by polyacrylamide gel electrophoresis and genotyping of the VP4 and VP7 alleles using well-established seminested multiplex reverse-transcription polymerase chain reaction methods. In 2003, 62% of strains exhibited the short rotavirus electropherotype, and the most common rotavirus strain was G2P[4]. In subsequent years, predominant rotavirus strains included G1P[8] and G1P[6] in 2004, G3P[8] and G3P[6] in 2005, and G1P[8] in 2006. For the 4 years of the study, rotavirus strains with P[6] genotype were detected in 25% of all rotavirus-positive specimens. In addition, unusual G12P[6] and G8 strains were detected at a low frequency. These results reflect the diversity of rotavirus strains circulating in South African communities.

Characterization of genotype G8 strains from Malawi, Kenya, and South Africa

Reviews of the global distribution of rotavirus genotypes have revealed the continuous circulation of G8 strains in Africa, often responsible for more cases of rotavirus disease than the more common G1–G4 rotavirus strains. During the study, genotype G8 strains from Malawi, Kenya, and South Africa were detected and the VP7 and VP4 genes of selected specimens were sequenced. Results indicated that G8 strains appeared to reassort frequently and were associated with P[6], P[4], and P[8] specificity. Phylogenetic analysis suggested that G8 strains occurred in a North/South African phylogenetic divide. In addition, G8 strains appear to be able to infect non‐human primates and strains with close phylogenetic relationships were detected in the same year on two continents. Any rotavirus vaccine introduced into African environments will need to demonstrate protective efficacy against unusual genotype combinations, new serotypes, and animal strains. Therefore, continuous monitoring of rotavirus strains in human and animal populations in Africa is a necessity. J. Med. Virol. 82:2073–2081, 2010.

Prospective hospital-based surveillance to estimate rotavirus disease burden in the Gauteng and North West Province of South Africa during 2003–2005

BACKGROUND Rotavirus is considered to be the most common cause of serious acute dehydrating diarrhea worldwide. However, there is a scarcity of information on rotavirus disease burden in sub-Saharan Africa. METHODS We conducted prospective, hospital-based surveillance for rotavirus diarrhea among children <5 years of age at the tertiary care Dr. George Mukhari Hospital (DGM) and at the Brits district Hospital (BH) in the Gauteng and North West Provinces in South Africa; we estimated that up to 80% of children <5 years of age in their catchment areas who are hospitalized for diarrhea are admitted to one of these hospitals. RESULTS At DGM, 2553 children <5 years of age were admitted for diarrhea from January 2003 through December 2005, and 852 children <5 years of age were treated for diarrhea at BH during 2004-2005. We examined stool specimens from 450 children (53%) at BH and from 1870 children (73%) admitted to DGM. An estimated 22.8% (95% confidence interval [CI], 21.2%-24.5%) of the children hospitalized with diarrhea at DGM were rotavirus positive, and the corresponding figure at BH was 18.2% (95% CI, 14.9%-22.1%). Among children <5 years of age admitted to DGM for any reason, an estimated 5.5% (95% CI, 5.1%-6.0%) had rotavirus diarrhea. Our incidence estimates suggest that 1 in 43-62 children in the area is likely to be hospitalized with rotavirus diarrhea by 2 years of age. CONCLUSIONS Prevention of serious rotavirus illness by vaccination will substantially reduce not only the disease burden among young children but also the case load in South African health care facilities.