In our double-blind, placebo-controlled, trial in the developing Asian countries of Bangladesh and Vietnam, pentavalent rotavirus vaccine was efficacious for prevention of severe rotavirus gastroenteritis in infants for nearly 2 years of follow-up. Moderately high vaccine efficacy in the first year of life led to a substantial reduction of severe rotavirus gastroenteritis. Efficacy was slightly reduced in the second year of life, but, combined with a lower background incidence of the disorder than was reported in the first year (table 3), still led to a significant reduction in incidence of severe rotavirus gastroenteritis.

Our study is the first clinical efficacy trial of an already licensed rotavirus vaccine in developing countries in Asia. Although pentavalent rotavirus vaccine significantly reduced the burden of severe disease, our reported efficacy measurements were lower than were those reported in trials of this vaccine in more industrialised countries in the USA, Europe, and Latin America.3,14—16 Although many factors could have led to the lower estimate of vaccine efficacy, our estimates might not be comparable with those reported in previous trials in the developed world. Studies were designed differently and used different clinical scoring systems. Furthermore, inclusion criteria were broad in this trial, and severity scores might have been less precise than in other trials because clinical measurement procedures were done only during stay in a medical facility, and relied on parental recall.

Higher efficacy point estimates were consistently measured in Vietnam than in Bangladesh, apart from efficacy against severe rotavirus gastroenteritis with a Vesikari score of 15 or more. These differences might be attributable to a low degree of precision in this study, because it was not designed to make statistical comparisons between countries. Alternatively, these differences could be a result of the different socioepidemiological circumstances of the study populations in the two countries’ study sites. For example, before the start of the study, the infant mortality rate in the study areas was two times higher in Bangladesh than it was in Vietnam (29·7 deaths per 1000 livebirths in Bangladesh17vs 14·7 deaths per 1000 livebirths in Vietnam18). Incidence of severe rotavirus gastroenteritis (Vesikari score ≥11) in the first year of life was substantially higher in Bangladesh than in Vietnam, suggesting a higher force of infection in Bangladesh. Equally, about 75% of children in developing countries have their first rotavirus infection before the age of 12 months,19 a time in their early childhood when they are most susceptible to diarrhoeal disease morbidity and mortality. Finally, the difference in rates of severe rotavirus gastroenteritis measured between the sites might have been attributable to different case-capture sensitivity and the health-care-seeking behaviours of the participants’ carer givers, although this should not have been different between vaccine and placebo groups.

Although there is no known immune correlate of protection for rotavirus, serum antirotavirus IgA response or serum neutralising antibody response might be a metric of vaccine efficacy.3, 14, 20 Serum IgA seroresponse and geometric mean titres after the third dose of vaccine or placebo were higher in children in Vietnam than in children in Bangladesh. Furthermore, IgA geometric mean titres measured in infants in Vietnam were similar to those in infants in Latin America,8 but IgA antibody concentrations measured in infants in Bangladesh were similar to those of impoverished populations in Africa.21 This difference might be because of dissimilar epidemiological and socioeconomic circumstances for infants in Bangladesh and Vietnam.

Geometric mean titres of serum neutralising antibodies to the human rotavirus serotypes contained in pentavalent rotavirus vaccine varied across a wide range (table 4), and were similar to those reported in children in Latin America8 (which ranged from 21·2 for serotype G3 to 125·8 for serotype G1). However, we noted a lower serum neutralising antibody seroresponse than has been reported previously.8 This low seroresponse was probably attributable to high concentrations of serum neutralising antibodies before the first dose of vaccine or placebo, because of high concentrations of residual maternal (transplacental or breast milk) antibodies. However, results from specific antibody tests (ie, IgA or serum neutralising antibody) should not be overinterpreted.

Several factors affect the immune response to live oral vaccines, including the concentration of transplacentally acquired maternal antibody, immune and non-immune components of breast milk, the amount of gastric acid in the digestive tract, micronutrient malnutrition, interfering gut flora, and diarrhoeal and immune system diseases.22 Additionally, oral poliovirus vaccine reduces the immunogenicity of live oral rotavirus vaccines when given concomitantly,8, 23, 24 and, unlike in previous efficacy trials of pentavalent rotavirus vaccine, around 90% of infants in this study received both vaccines concomitantly. Infants in Bangladesh might, in fact, have lower immune responses to pentavalent rotavirus vaccine than do infants in Vietnam, which puts them closer to an unknown minimum threshold for protection. Understanding of the contribution of these factors to low immune responses and efficacy in specific populations might allow investigators to design immunisation programmes and vaccines that will be more effective for prevention of severe rotavirus gastroenteritis than those we have at present.

Two rotavirus vaccines, RotaTeq and Rotarix (GlaxoSmithKline, Rixensart, Belgium), have been approved for use by national regulatory authorities in many countries worldwide. However, until efficacy results of trials from representative regional populations were available, neither vaccine could be incorporated into the routine public health immunisation programmes in developing countries in Asia and Africa that were eligible to receive GAVI Alliance co-funding. In November 2007, WHO convened an ad-hoc group of experts who concluded that efficacy data could be extrapolated to populations that are in equivalent child-mortality strata,25 thus helping to alleviate the need for regional data. In 2009, results were presented to WHO’s SAGE from an efficacy trial26 in Malawi and South Africa that showed that Rotarix was efficacious against severe rotavirus gastroenteritis in the first year of life. These results, along with postmarketing effectiveness data for RotaTeq from Nicaragua27 and the USA,28 and data for Rotarix from El Salvador29, led the WHO group to expand recommendations for rotavirus vaccination to all regions of the world.30

With a WHO recommendation for rotavirus vaccines now in place,31 governments of developing countries in Africa and Asia are deciding how to prioritise introduction of rotavirus vaccine in their public health agendas. Our trial shows that a live oral rotavirus vaccine has the potential to halve the incidence of severe rotavirus gastroenteritis in developing populations in Asia. Alongside efficacy results for this vaccine in Africa,21 our study supports WHO’s strong recommendation for expansion of rotavirus vaccine use to the poorest nations in Africa and Asia. Rotavirus vaccines have the potential to protect the lives of nearly 2 million children in the next decade alone.32

KZ, DDA, JCV, SS, MY, MJD, VDT, SPL, MLC, DAS, FS, ADS, KMN, and MC contributed to the study design. KZ, DDA, SS, MY, GP, VDT, LPM, and LHT contributed to the implementation of the study and supervision at the sites. JCV, MJD, KL, SBR, FS, ADS, KMN, and MC contributed to planning of protocol-stated analyses and post-hoc analyses. MJD designed and did the statistical analysis and verified its accuracy. MJD, SBR, and MC contributed to compiling of the official clinical study report. KZ, DDA, JCV, SS, MY, MJD, KL, ADS, KMN, and MC took part in a 2 day meeting to discuss and interpret the data. KZ was the principal investigator for Bangladesh; DDA was the principal investigator for Vietnam. KN led the clinical team at PATH; MC led the clinical team at Merck. All authors had full access to the data. KZ, DDA, JCV, SS, MY, SPL, MLC, KL, SBR, DAS, FS, ADS, KMN, and MC helped draft this report or critically revise the draft. All authors reviewed and approved the final version of the report.

MJD, MLC, SBR, FS, and MC are employees of Merck and own shares in the company. DAS was Director of ICDDR,B at the time of initiation of the study; after his departure from ICDDR,B, he received consultancy fees as part of his ongoing participation in the site’s conduct of the study. Consultancy fees were part of the ICDDR,B budget, which was funded by PATH. ICDDR,B, NIHE, and IVI were funded by PATH’s Rotavirus Vaccine Program through a GAVI Alliance Grant to PATH to conduct this trial. All other authors declare that they have no conflicts of interest.

The study, with protocol V260-015, was designed, managed, undertaken, and analysed by the co-sponsors in collaboration with the site investigators and under the supervision and advice of the data and safety monitoring board. Investigators and their institutions were funded by PATH’s Rotavirus Vaccine Program, with a grant from the GAVI Alliance. We thank the volunteers and their families; the scientific advisers Shams El Arifeen and Tasnim Azim (International Centre for Diarrhoeal Disease Research, Bangladesh); participating investigators Abu Syed Golam Faruque, Al Fazal Khan, and Ilias Hossain (International Centre for Diarrhoeal Disease Research, Bangladesh); all other staff from the International Centre for Diarrhoeal Disease Research, Bangladesh; the scientific advisers John Clemens and Paul Kilgore (International Vaccine Institute in Korea) and Truong Tan Minh (Khanh Hoa Health Service), and participating investigators Nguyen Hien Anh (National Institute of Hygiene and Epidemiology) and Phu Quoc Viet (Khanh Hoa Health Service); staff from the Department of Pediatrics, Khanh Hoa General Hospital, from the Pasteur Institute in Nha Trang, and from the 16 participating Commune Health Centres in Nha Trang; the members of the data and safety monitoring board (King Holmes [Chairman], Wasif Ali Khan, Edward Tsiri Agbenyega, Grace Irimu, Mamadou Marouf Keita, Dinh Sy Hien, Nik Zarifah, Nik Hussain Reed, and Janet Wittes); Penny M Heaton and Michelle G Goveia for their contribution to the design of the study; Bradley Raybold for contributions to the initiation and implementation of the study; Fay DiCandilo and Margaret Nelson for contributing to careful review of the data; Donna Hyatt for data management; Laura Mallette and Vladimir Liska for laboratory data coordination; Richard Ward and Monica McNeal for overseeing laboratory assays; Tracy Burke for ensuring adequate vaccine and placebo supplies; Family Health International and PharmaLink, especially Carolyn Enloe, Vivian Bragg, Laura Niver, Jen Auerbach, Linda McNeil, and all the Family Health International field monitors and safety-reporting staff; Vu Minh Huong (PATH) for monitoring assistance; Joyce Erickson (PATH) for contracting and financial analysis; and Carolien Bakker and David Oxley (PATH) for administrative assistance.

source: LANCET