ARTICLE |
a Department of Infectious Diseases, Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubirán, Mexico City, Mexico
b GlaxoSmithKline Mexico, Mexico City, Mexico
c GlaxoSmithKline Latin America and Caribbean, Rio de Janeiro, Brazil
d GlaxoSmithKline Biologicals, Rixensart, Belgium
| ABSTRACT |
|---|
|
|
|---|
METHODS. From June 2001 through May 2003, 405 healthy infants were randomly assigned to 1 of 3 vaccine groups (virus concentrations 104.7, 105.2, and 105.8 infectious units) and to a placebo group and were monitored to the age of 2 years. The vaccine/placebo was administered concurrently with diphtheria-tetanus toxoid-pertussis/hepatitis B/Haemophilus influenzae type b vaccine at 2 and 4 months of age. After the administration of the first vaccine/placebo dose, weekly home visits to collect information regarding infant health were conducted. Stool samples were collected during each gastroenteritis episode and tested for rotavirus antigen and serotype.
RESULTS. The vaccine was well tolerated and induced a greater rate of seroconversion than observed in infants who received placebo. For the pooled vaccine groups, efficacy after 2 oral doses was 80% and 95% against any and severe rotavirus gastroenteritis, respectively. Efficacy was 100% against severe rotavirus gastroenteritis and 70% against severe gastroenteritis of any cause with the vaccine at the highest virus concentration (105.8 infectious units). The predominant infecting rotavirus serotype in this cohort was wild-type G1 (85%). Adverse events, including fever, irritability, loss of appetite, cough, diarrhea, and vomiting, were similar among vaccinees and placebo recipients.
CONCLUSION. This new oral, live, attenuated human rotavirus vaccine was safe, immunogenic, and highly efficacious in preventing any and, more importantly, severe rotavirus gastroenteritis in healthy infants. This vaccine produced comparable protection to natural infection.
Key Words: vaccines rotavirus diarrhea viral gastroenteritis immunization
Abbreviations: HRV—human rotavirus vaccine IU—infectious units CI—confidence interval IgA—immunoglobulin A ELISA—enzyme-linked immunosorbent assay ITT—intention-to-treat ATP—according-to-protocol
Rotavirus is the leading cause of severe diarrhea and dehydration among children in both developed and developing countries.1–5 Despite the recognized efficacy of oral rehydration in treating dehydrating diarrhea, each year severe rotavirus gastroenteritis causes 350000 to 600000 deaths in children who are younger than 5 years.1 It also accounts for 2 million childhood hospital admissions with an estimated cost of more than $1 billion per year.5 Because rotavirus disease remains widely prevalent in industrialized countries, it is unlikely that the improvement of hygienic and sanitary conditions alone will have a dramatic impact on the disease in developing countries. Therefore, effective immunization against rotavirus has been proposed as the most cost-effective intervention to reduce the disease burden that is associated with this infection in infants and young children worldwide.
The greatest fatal consequences of rotavirus gastroenteritis are seen in infants and children from developing countries, where access to medical care is limited.2–4 In Mexico, diarrheal diseases remain a major public health problem and are the second cause of death among children.6 Epidemiologic surveillance has shown that most diarrhea-associated deaths and hospital admissions in Mexican children occur during the fall and winter seasons, when the highest frequencies of rotavirus infections are observed.7–12
In 1998, the first rhesus and human reassortant rotavirus vaccine was licensed in the United States (RotaShield; Wyeth Lederle Laboratories, Philadelphia, PA).13 However, the vaccine was withdrawn within 1 year because of its association with intussusception.14 The withdrawal of this vaccine stimulated the development of alternative safe and effective vaccines that could be made available to children worldwide. Among the new vaccine candidates, a live, attenuated, monovalent G1[P8], human rotavirus vaccine (HRV) was developed. Initial studies on the 89-12 vaccine demonstrated its safety, immunogenicity, and efficacy in infants.15,16 GlaxoSmithKline Biologicals (Rixensart, Belgium) implemented several process changes to the 89-12 vaccine candidate to develop a lyophilized vaccine that contains a genetically stable cloned 89-12 HRV strain RIX4414. A phase IIb, double-blind, placebo-controlled, randomized clinical study was conducted to determine the optimal viral concentration of the RIX4414 live, attenuated HRV and to assess its efficacy and safety in Brazil, Mexico, and Venezuela.17 We report the results in Mexican infants and review the efficacy and immunogenicity of this HRV with respect to our previous studies on the natural protection against rotavirus infection.8–10
| METHODS |
|---|
|
|
|---|
Study Area
The study was conducted in San Pedro Mártir, the same setting in Mexico where natural protection against subsequent rotavirus infection was studied.8,9,18 In this highly endemic area for diarrheal diseases, a cohort study of 200 children demonstrated that rotavirus infections are highly prevalent (60 episodes per 100 child-years; 95% confidence interval [CI]: 56–64) and that half of the children infected will develop diarrhea (30 episodes per 100 child-years; 95% CI: 25–34), with the highest frequency and severity occurring in infants 4 to 6 months of age and during the winter season.8,9
Study Population
A census of all pregnant women and infants who were younger than 12 weeks was conducted in the area at the beginning and throughout the enrollment period. A prospective consecutive sample of 405 term, healthy infants who were 6 to 12 weeks of age at the time of the first dose of the study vaccination course and whose mothers were willing to participate and remain in the study area were included. Children were excluded when they had birth defects, had an underlying disease, were immunocompromised or had contact with a pregnant woman or with an immunosuppressed individual, or were participating in another investigation. The study was approved by the institutional review board of the National Institute of Medical Sciences and Nutrition in Mexico City. Written informed consent was obtained from parents.
Randomization
A randomization list was generated using a standard Statistical Analysis System (SAS Institute, Cary, NC) program to number vaccine doses. Eligible infants were randomly assigned to 1 of the 4 study groups (3 vaccine groups and a placebo group) by study physicians. A randomization blocking scheme (1:1:1:1 ratio) was used to ensure balance between treatments. Blinding was maintained until study termination.
Vaccination
Routine vaccinations were administered following recommendations of the Mexican Expanded Program of Immunization with the exception of oral polio vaccine, which was given at least 14 days before or after administration of the study vaccine. The HRV and placebo were prepared by reconstituting the lyophilized product with the supplied diluent (calcium carbonate buffer). Each reconstituted dose (1 mL) consisted of 104.7, 105.2, or 105.8 IU of HRV, strain RIX4414 (GlaxoSmithKline Biologicals). The placebo had the same constituents as the active vaccine but without the vaccine virus. Children received orally 2 doses of the reconstituted vaccine or placebo at 2 and 4 months of age. Children were closely observed for at least 30 minutes to identify any reaction after vaccine administration. There were no restrictions on infant feeding.
Study Surveillance
All children enrolled were followed prospectively until 1 year of age. Efficacy surveillance started on day 0 (administration of vaccine/placebo dose 1) and continued until 1 year of age. A subset of 272 (67%) of 405 children who completed the first efficacy period and whose parents voluntarily accepted and signed an informed consent participated in the second efficacy follow-up period (from 1 to 2 years of age or end of the 2003 rotavirus season in May). Follow-up, sample collection, and testing were consistent with study procedures used in our previous cohorts.8,9 Briefly, field workers visited all households once a week to identify gastroenteritis cases and to collect diarrhea stool specimens throughout the 2-year study period. After the report of diarrhea, a study physician made a home visit to examine the child and obtain information about the occurrence of fever, vomiting, dehydration, and other complications. Diary cards were completed by parents or guardians during each diarrhea episode until symptoms resolved.
Caregivers recorded on reactogenicity diary cards the occurrence of fever, irritability/fussiness, loss of appetite, cough, vomiting, and diarrhea during the 15 days after vaccination. Field workers verified that the card was properly completed and collected it. At each visit, all adverse events observed by the study personnel or spontaneously reported by caregivers were evaluated. Caregivers were informed about signs of intussusception, such as the occurrence of severe abdominal pain, persistent vomiting, bloody stools, abdominal bloating, and fever up to 41°C, and they were asked to contact the investigator immediately if intussusception was suspected.
Stool and Blood Samples
Stool was collected on the first 72 hours of a gastroenteritis episode from all children.8,9 A convenience sample that included the first 252 (62%) of 405 infants enrolled also provided stool specimens on day 0 and day 7 after each dose to monitor for shedding of vaccine and wild-type rotavirus strains. Blood samples were collected from 386 (95%) of 405 infants immediately before the first vaccination and in 260 (64%) of 405 at 1 year of age for antirotavirus immunoglobulin A (IgA) serology. For determination of immune response to the vaccine, a convenience sample of the first 158 (39%) enrolled infants provided blood samples 2 months after the first dose, and 154 (38%) provided blood samples 2 months after the second dose. Serum samples were stored at –20°C until testing.
Rotavirus Detection and Serotyping
Stool samples were tested for rotavirus antigen by enzyme-linked immunosorbent assay (ELISA; Rotaclone; Meridian Diagnostics, Inc, Cincinnati, Ohio) and for other bacterial enteropathogens.19 All rotavirus positive stool samples were tested by reverse transcriptase–polymerase chain reaction to determine the G type.20 When any G1 rotavirus was detected in stool, vaccine virus was differentiated from wild-type rotavirus by sequence analysis.
Rotavirus Serology
Serum anti-rotavirus IgA antibodies were measured by a solid-phase ELISA10,13 and expressed in units per milliliter. The assay cutoff point for seropositivity was 20 U/mL.
Study Definitions
Diarrhea was defined as the presence of 3 or more looser-than-normal stools in 24 hours. A new diarrhea episode was separated from a previous episode by
5 days of being well. Rotavirus diarrhea was defined as detection of rotavirus antigen in a diarrheal stool specimen. Severity of diarrhea was assessed using Ruuska and Vesikari's13,21 score: <7 = mild, 7 to 10 = moderate, and
11 = severe. Serious adverse events were defined as any medical condition that resulted in death or persistent or significant disability, was life-threatening, or required prolonged hospitalization.
The primary end points were immunogenicity and efficacy against any and severe rotavirus gastroenteritis at 1 year of age. Secondary end points were efficacy after a second rotavirus season in children during the second year of life and reactogenicity to the study vaccine.
Statistical Analysis
Remote data entry was done using electronic case report forms on IBM-compatible microcomputers. Data were analyzed using SAS 8.2 and Pro StatXact 5 (Cytel Inc, Cambridge, MA). Intention-to-treat (ITT = total number of children vaccinated) and according-to-protocol (ATP) analyses were done. Incidence rates were calculated by dividing the number of rotavirus gastroenteritis of any and severe intensity by the number of child-years of observation by study group. Vaccine efficacy (incidence in controls – incidence in vaccinees/incidence in controls) was calculated with their corresponding 95% CIs.22
Role of the Funding Source
The study sponsors participated in study design, data analysis, and report review. They did not participate in data collection or data interpretation or in the decision to submit this article for publication.
| RESULTS |
|---|
|
|
|---|
|
|
|
Compared with the placebo group, vaccinated children had significantly fewer rotavirus gastroenteritis episodes of any severity (pooled incidence 6 episodes per 100 child-years vs 25 episodes per 100 child-years in the placebo group; P = .0001; Table 3). Efficacy of 2 vaccine doses against rotavirus gastroenteritis of any severity was 80% (95% CI: 46.3–93.5) for the pooled vaccine groups (ITT = 76.3%; 95% CI: 48.9–89.3). The vaccine was more effective in preventing severe rotavirus gastroenteritis (P < .001), with an efficacy of 95% (95% CI: 61.0–99.8) for the pooled vaccine groups (ITT = 90%; 95% CI: 65.3–98.3) and 100% for the intermediate and higher dosages. No child was hospitalized as a result of rotavirus gastroenteritis.
|
Efficacy During 2 Consecutive Follow-up Periods
The predominant rotavirus type was also G1 (85%). In addition, G4 and G9 types were identified. During the 2 consecutive follow-up periods, there was a statistically significant decrease in the percentage of children who were reported to have any (8.4% vs 20.6%; efficacy: 59.1%; 95% CI: 21.6–78.5; P < .003) or severe (0.4% vs 10.3%; efficacy: 96.4%; 95% CI: 75.0–99.9; P < .001) rotavirus gastroenteritis in the pooled HRV groups as compared with the placebo group.
Rotavirus Shedding
Vaccine virus shedding was observed in 22 (43%), 9 (21%), and 27 (61%) infants on the seventh day after the first dose and in 4 (10.5%), 4 (12%), and 5 (14.3%) infants on the seventh day after the second dose in the 104.7-IU, 105.2-IU, and 105.8-IU groups, respectively. None, except for 1 infant in the 105.8-IU vaccine group, excreted the vaccine strain 60 days after dose 1, immediately before dose 2. Among placebo recipients, none shed any rotavirus after the first or second dose.
Reactogenicity and Safety
All 405 infants who had received at least 1 dose of the study vaccine or placebo were included in the primary analysis of safety. Incidences of fever, diarrhea, vomiting, irritability, loss of appetite, and cough/runny nose reported during 15 days after any dose were similar among the 4 study groups (Fig 2). Overlapping 95% CI between each vaccine and placebo groups suggested that there was no increase in reactogenicity either with subsequent doses or with higher viral concentrations.
|
| DISCUSSION |
|---|
|
|
|---|
Longitudinal studies around the world have also shown that early and repeated rotavirus infections confer natural immunity in children and protect them against subsequent severe disease.23–25 In the Mexican cohort, children with an IgA titer >1:800, achieved early or after 2 consecutive symptomatic or asymptomatic rotavirus infections, had a lower risk for infection (adjusted relative risk: 0.21; P < .001) and diarrhea (adjusted relative risk: 0.16; P = .01) and were completely protected against moderate to severe diarrhea.10 After the first dose, the vaccine elicited an immune response similar to that induced by a first natural rotavirus infection. Higher anti-rotavirus IgA seroconversion rates were achieved after the second dose of the vaccine.
The 105.8-IU vaccine dosage resulted in the highest seroconversion rates, was highly efficacious against any and severe rotavirus diarrhea, and was more protective against diarrhea of any cause when compared with lower dosages. Therefore, subsequent phase III clinical trials were done with dosages of up to 106.5 IU.26 Production cost and the stability of vaccine titers to expiration date determined that the licensed product, Rotarix, should contain no less than 106 median cell-culture infective dose of the RIX4414 vaccine strain.
The RIX4414 vaccine was well tolerated, and there were no serious adverse events associated with vaccination. Although the study's sample size was too small to draw any conclusions regarding the risk for intussusceptions, certainly the larger phase III multinational study26 proved the safety and confirmed the vaccine's efficacy.
In this study, it was not possible to measure the impact of the vaccine in the prevention of very severe gastroenteritis, because no rotavirus-associated hospitalizations or deaths were observed in the placebo or vaccine groups, probably as a result of the close monitoring and contact with health workers and because of greater use and early administration of rehydration therapy, which may have altered the course of the illnesses. It is also possible that study populations were more aware about the risk for this illness and acted in consequence. Nevertheless, important health implications can be derived from these observations: (1) by grading the severity of the diarrhea episodes, we proved the high efficacy of the RIX4414 vaccine in preventing any and severe disease; (2) the vaccine could be even more beneficial in populations in which there is no awareness of the disease, there is lack of early and appropriate access to medical care, and close monitoring of children who are at high risk for severe disease is almost impossible, although studies in less privileged populations will be necessary to prove this concept.
The withdrawal of the first rotavirus vaccine, RotaShield, taught important lessons and provided directions for the development of the next generation of rotavirus vaccines.27 A vaccine that may prevent 440000 childhood deaths each year, or 1 in 20 deaths among children younger than 5 years, is now a reality and will become available sooner than ever anticipated in the developing world, where they are most needed.
| CONCLUSIONS |
|---|
|
|
|---|
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
Address correspondence to Guillermo M. Ruiz-Palacios, MD, Department of Infectious Diseases, Instituto Nacional de Ciencias Médicas y Nutrición, Salvador Zubirán Vasco de Quiroga 15, Tlalpan 14000, Mexico. E-mail: gmrps{at}servidor.unam.mx
Financial Disclosure: Drs Cervantes, Costa-Clemens, and DeVos are employees of GlaxoSmithKline, and this study was sponsored by GlaxoSmithKline Biologicals.
This work was presented in part at the meeting of the World Society for Pediatric Infectious Diseases; November 19–23, 2002; Santiago, Chile.
| REFERENCES |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||