 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
ARTICLE |
a Kentucky Pediatric/Adult Research, Bardstown, Kentucky
b University of Tampere Medical School, Tampere, Finland
c Merck Research Laboratories, North Wales, Pennsylvania
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
OBJECTIVE. Our goal was to evaluate the efficacy, immunogenicity, and safety of pentavalent rotavirus vaccine at the end of shelf life in healthy infants.
PATIENTS AND METHODS. During 2002–2004, 1312 healthy infants 
6 to 12 weeks old from the United States (47%) and Finland (53%) were randomly assigned to receive 3 oral doses of vaccine (vaccine at 1.1 x 107 infectious U per dose) or placebo 4 to 10 weeks apart. Infants were to be followed for acute gastroenteritis through 1 rotavirus season after vaccination and for adverse events postvaccination.
RESULTS. Three doses of pentavalent rotavirus vaccine at the end of shelf life demonstrated efficacy against rotavirus gastroenteritis caused by human G-serotypes included in the vaccine (G1–G4). Efficacy against severe rotavirus gastroenteritis was 100%, and efficacy against any rotavirus gastroenteritis regardless of severity was 72.5%. A threefold rise in G1 serum neutralizing was observed in 57% and in anti-rotavirus immunoglobulin A in 96% of pentavalent rotavirus vaccine recipients. No statistically significant increase in vomiting, diarrhea, or irritability was observed among pentavalent rotavirus vaccine recipients compared with placebo recipients within the 7-day period from each dose. A statistically significant increase in fevers (
100.5°F, rectal equivalent) was observed among pentavalent rotavirus vaccine recipients compared with placebo recipients after dose 1.
CONCLUSIONS. This pentavalent human-bovine rotavirus vaccine was generally well tolerated, efficacious, and immunogenic at the end of shelf life.
Key Words: gastroenteritis • rotavirus vaccine • rotavirus • safety • efficacy
Abbreviations: RVGE—rotavirus gastroenteritis • WC3—Wistar calf 3 strain of bovine rotavirus • PRV—pentavalent rotavirus vaccine • REST—Rotavirus Efficacy and Safety Trial • EIA—enzyme immunoassay • RT-PCR—reverse-transcriptase polymerase chain reaction • AGE—acute gastroenteritis • SAE—serious adverse event • IgA—immunoglobulin A • CL—confidence limit • SIDS—sudden infant death syndrome
Rotavirus is the leading cause of severe acute gastroenteritis in children under the age of 5 years worldwide.1 In the United States, rotavirus gastroenteritis (RVGE) rarely causes mortality, but it does cause significant morbidity.2,3 Recent hospital-based studies in the United States estimate that RVGE accounts for 90000 hospitalizations annually, which is nearly half of all hospitalizations for diarrhea among children <5 years old.4,5 In Finland, RVGE also places a significant burden on the health care system where 1 in 33 children under 5 years of age requires hospitalization.6
A safe and effective rotavirus vaccine is a high priority because of the extraordinary medical burden of RVGE, its impact on public health resources, and the lack of alternative measures to control the disease. The most prevalent human serotypes are G1, G3, G4, and G9 in conjunction with P[8] and G2 in conjunction with P[4]. These strains represent 88% of rotavirus strains worldwide.7 Thus, a pentavalent (G1–G4, and P[8]) human-bovine (WC3) reassortant rotavirus vaccine (PRV) was developed (RotaTeq, Merck & Co, Inc, Whitehouse Station, NJ), using WC3 bovine rotavirus as a backbone.8
In a Phase II dose-ranging clinical study, which evaluated several formulations and various potencies of the vaccine, the pentavalent formulation with a middose potency demonstrated the highest efficacy estimate and a robust immune response.9 These data were the basis for the selection of the PRV intended for licensure, including the vaccine evaluated in the large-scale Rotavirus Efficacy and Safety Trial (REST).10 Our study confirms the data obtained from the REST, where the vaccine potency was higher, as it assesses the efficacy, immunogenicity, and safety of PRV at the end of its shelf-life potency. Because PRV is a live-virus vaccine, in which the virus decays over time, this study was designed to evaluate the vaccine at the end of its 24-month shelf life.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Healthy infants between 6 to 12 weeks of age were eligible for the study. Infants were to be excluded if they had or were to receive oral poliovirus vaccine at any time during the study or in the 42 days before the first dose. Administration of other licensed childhood vaccines was permitted. In the United States, 80% of infants received the first study vaccination on the same day as the licensed childhood vaccinations. In Finland, study sites were separated from health care clinics; thus, the first study vaccination was typically 2 weeks before administration of licensed childhood vaccinations.
PRV contained 5 human-bovine reassortant rotaviruses of the WC3 (bovine G6) strain, consisting of 4 VP7 reassortants (human G1–G4) and 1 VP4 reassortant (human P[8]). The aggregate titer of the vaccine was 1.1 x 107 infectious units per dose. On the basis of stability data, the potency of vaccine was consistent throughout the duration of the study. Enrolled infants were randomly assigned 1:1 by using computer-generated allocation schedules to receive either vaccine or visibly indistinguishable placebo in a sucrose-citrate buffer administered orally as three 2-mL doses 4 to 10 weeks apart. The placebo was identical to the vaccine except that it did not contain the rotavirus reassortants or trace trypsin. There were no restrictions for feeding before or after vaccination.
The study was initially conducted in the Northeastern, Southern, and Midwestern regions of the United States and in Finland through 1 rotavirus season (2002–2003). However, because enrollment was slower than expected, a second cohort of infants was enrolled (2003–2004) at study sites in the Western region of the United States and Finland, who were also followed through 1 rotavirus season.
Midway through enrollment of the second cohort, based on blinded data from the trial, fewer cases of RVGE occurred during the 2002–2003 rotavirus season than originally projected. Thus, the sample size was increased up to 1400 infants to obtain enough cases to have adequate statistical power to evaluate the efficacy of the vaccine.
Evaluation of Efficacy
The primary end point of the study was vaccine efficacy against naturally occurring RVGE of any severity caused by a G-serotype included in PRV (G1–G4) through 1 rotavirus season after vaccination. Efficacy against moderate-and-severe and severe RVGE was also evaluated. Efficacy results from both cohorts were combined for the efficacy analyses.
A case of RVGE was defined as meeting both of the following criteria: (a) 3 watery or looser than normal stools within a 24-hour period and/or forceful vomiting; and (b) rotavirus antigen detection by enzyme immunoassay (EIA) in the stool sample. The primary analysis of efficacy included only cases caused by naturally occurring rotavirus of serotypes G1, G2, G3, or G4 as confirmed by reverse-transcriptase polymerase chain reaction (RT-PCR) occurring at least 14 days after the third dose of PRV and/or placebo. Ideally, 2 stool samples were to be collected: the first within 24 hours of symptom onset and a second sample 24 hours later; stool samples were required by protocol to be collected within 14 days of symptom onset.
Active surveillance for RVGE was conducted during the rotavirus season with contacts every 2 weeks with the parent or guardian. The rotavirus season was prospectively determined by using historical epidemiologic data for each region.11
To determine whether a potential acute gastroenteritis episode (AGE) met the clinical case definition and to quantify the severity of the episode, parents or guardians were instructed to complete a diary card documenting the infant's clinical gastrointestinal symptoms until resolution. A validated clinical scoring system based on the intensity and duration of symptoms of fever, vomiting, diarrhea, and behavior changes was used to rate the severity of the AGE. Each episode was graded on a 24-point scale, where a score 
8 was designated as mild, >8 was designated as moderate-and-severe, and >16 was designated as severe disease.10,12
Stool samples were tested for rotavirus antigen by EIA at Cincinnati Children's Hospital Medical Center (Cincinnati, OH).13 If the EIA was positive, the serotype was identified by RT-PCR at Merck Research Laboratories (Wayne, PA). All rotavirus EIA-positive stools, collected at any time throughout the study, were further evaluated for shedding of viable vaccine reassortants by plaque assay and RNA electropherotyping at Children's Hospital of Philadelphia (Philadelphia, PA).14,15
Evaluation of Immunogenicity
Approximately 1 mL of serum was obtained from a subset of infants at prespecified sites immediately before the first vaccination and 42 days after the third vaccination. Prevaccination and postvaccination sera were analyzed for serotype-specific rotavirus neutralizing antibody against human serotypes G1-G4, P[8], G6 and P[5] (the WC3 parent bovine strain has a G6 serotype and P[5] genotype), and for serum anti-rotavirus immunoglobulin A (IgA). Assays were validated and performed at Cincinnati Children's Hospital Medical Center. Seroconversion was defined as a threefold or greater rise in antibody titer from baseline to postvaccination (Merck, written communication, 1998–2001).16
Safety Surveillance
All infants were followed for clinical adverse events, including intussusception, for 42 days after each vaccination. Serious adverse events (SAEs) were to be reported immediately, which included intussusception and hospitalizations. Parents or guardians were provided diary cards and were instructed to record daily temperatures for the infant for 7 days after each vaccination. The protocol defined an elevated temperature as 100.5°F rectal equivalent (axillary temperatures were converted to rectal equivalent by adding 2°F). The parent/guardian was also to record the number of episodes of vomiting and diarrhea and to document any behavioral changes for the first 7 days after vaccination.
Potential cases of intussusception were adjudicated by an independent blinded committee. After each case was adjudicated, a data and safety-monitoring board unblinded the treatment arm and made recommendations about continuing all ongoing PRV clinical trials. Throughout the adjudication and safety monitoring process, all study personnel remained blinded to the treatment arm and adjudication results of the potential intussusception cases.
Parents or guardians were contacted by the study site on day 7, day 14, and day 42 after each vaccination and asked about SAEs. Data on cases of intussusception, deaths, or other SAEs determined to be vaccine-related by the investigator were collected throughout the study.
Statistical Methods
Analysis of Efficacy
The primary null hypothesis was that the vaccine efficacy against G1-, G2-, G3-, or G4-specific cases of RVGE occurring at least 14 days after dose 3 through 1 rotavirus season would be 0%. This hypothesis was tested by using an exact binomial procedure based on the proportion of infants with RVGE in the PRV group, conditional on all infants with RVGE. An exact 95% confidence interval was also computed.
The primary efficacy analysis was based on the per-protocol population, which consisted of infants not violating the protocol (eg, received all 3 doses of study material and received study material that was not involved in a temperature excursion), and used the per-protocol case definition. Secondary analyses based on the per-protocol population examined efficacy against moderate-and-severe disease and exclusively severe disease, and efficacy against all EIA-positive cases. Additional efficacy analyses were based on an intention-to-treat population, which consisted of all infants regardless of protocol violations who received 1 dose. These analyses used 2 different case definitions: one that included all cases as defined by the per-protocol case definition occurring any time after the first dose, and a second that included cases starting at least 14 days after the first dose. For all efficacy analyses, infants with multiple episodes that met the case definition were counted only once as a positive case.
Analysis of Immunogenicity
Immunogenicity was evaluated in a subset of 150 infants that included the initial 100 infants from the first cohort and the initial 50 infants from the second cohort. To evaluate immunogenicity, the proportion of infants achieving a threefold rise or greater from baseline to after the third vaccination in titer, and geometric mean titers as measured by serum neutralizing antibody responses to G1, G2, G3, G4, P[8], G6, and P[5] and serum anti-rotavirus IgA were summarized by treatment group in the per-protocol population. Infants whose stool was rotavirus EIA-positive were excluded from the immunogenicity analysis.
Analysis of Safety
The safety hypothesis was that the PRV would be generally safe and well tolerated with respect to all adverse events. Adverse events occurring within 42 days of any dose were summarized as frequencies and percentages by treatment group and by type of event, along with associated risk differences and exact 95% confidence intervals. Within 7 days of each dose, the following adverse events were evaluated by also using 2-sided P values: elevated temperatures (100.5°F, rectal equivalent), vomiting, diarrhea, and irritability. All infants with safety follow-up were included in the safety summaries.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Overall, 1115 infants (551 and 564 infants in the PRV and placebo groups, respectively) were assessed for the primary efficacy analysis. A total of 99 infants in the PRV group (15.2%) and 96 infants in the placebo group (14.5%) were excluded from the primary efficacy analysis because of protocol violations (n = 66 and 61, respectively) or inadequate data with respect to the per-protocol case definition (n = 33 and 35, respectively).
A total of 69 infants (15 PRV recipients) met the case definition of RVGE occurring at least 14 days after the third dose of vaccine or placebo caused by serotypes G1, G2, G3, and G4: 66 infants (13 PRV recipients) were infected with a G1 serotype, and 3 (2 PRV recipients) were infected with a G3 serotype. The associated efficacy against any severity was 72.5% (95% confidence limits [CLs]: 50.6%, 85.6%) through 1 rotavirus season after vaccination (Table 2).
|
Among all infants who received at least 1 dose, by using the per-protocol case definition for RVGE cases occurring anytime after the first dose, an estimate of efficacy was 58% (based on a case split of 27 PRV: 64 placebo cases). Of these cases, 21 in the PRV group and 63 in the placebo group occurred at least 14 days after the first dose, for an efficacy estimate of 67%. A posthoc evaluation of the VP7 sequence of EIA-positive stool samples showed that among the 27 PRV cases, 18 were a wild-type strain (providing an efficacy estimate of 75%). Nine cases were associated with plaque-negative vaccine-strains, indicating shedding of degraded, nonviable, vaccine virus. Of the 9 cases, 4 occurred within 14 days of the first dose, 2 occurred at least 14 days from the first dose, another 2 occurred within 14 days of the second dose, and the final case occurred at least 14 days from the second dose.
Immunogenicity
Among the subset of infants in which serum specimens were obtained, the percentage of infants who achieved at least a threefold rise in G1 serum neutralizing antibody titers from baseline to after dose 3 was 56.7% among PRV recipients (N = 67) and 2.7% among placebo recipients (N = 73). In addition, 95.5% of PRV recipients (N = 67) and 12.3% of placebo recipients (N = 73) had at least a threefold rise in serum anti-rotavirus IgA. The proportion of infants with at least a threefold rise in SNA titers varied by serotype being tested (Fig 1).
|
100.5°F, rectal equivalent) after dose 1 was seen in PRV recipients when compared with placebo recipients (13.4% vs 8.8%, P = .01). During the week after the second or third vaccinations, no statistically significant increase in fever was observed. There were only 2 (0.3%) infants in each of the PRV and placebo groups that reported a high fever (102.5°F, rectal equivalent) after dose 1.
Approximately 80% and 99% of children reported rectal temperatures in the United States and Finland, respectively. Figure 2 shows the rates of fever (100.5°F, rectal equivalent) by country for each of the days during the week from the first vaccination. The rate of fever was 1.5-fold higher in PRV recipients than in placebo recipients. Among infants in the United States, who typically received routine childhood vaccinations at the same time, 25.7% of PRV recipients and 17.7% of placebo recipients reported a fever following dose 1. Among infants in Finland, where the majority did not receive routine childhood vaccinations at the same time as the study vaccination, 4.0% of PRV recipients and 1.5% of placebo recipients reported a fever following dose 1.
|
Serious Adverse Events
SAEs were reported in 48 infants (3.7%); among these infants, 21 (43.8%) and 27 (56.2%) received PRV and placebo, respectively (Table 3). All infants except for 2 were hospitalized. No cases of intussusception were reported during the course of this study. One death occurred in a Hispanic male who received PRV. The infant died from presumed sudden infant death syndrome (SIDS) at almost 5 months of age, which was 6 weeks after receiving dose 2. It was noted on the day of death that the infant had a history of an upper respiratory infection for 2 days. The death was assessed by the blinded investigator as not vaccine-related. Six infants (1 infant received PRV and 5 infants received placebo) discontinued from the study because of a serious clinical adverse event; none were reported to be vaccine-related. There were 2 infants hospitalized with fever within the 42 days from vaccination. One infant developed fever 4 days after dose 1 along with an urinary tract infection, and the other infant developed fever 36 days after dose 1 and was reported to have concurrent rhinorrhea. Neither event was considered to be vaccine-related.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The primary circulating rotavirus strain during the study period was G1, limiting the assessment of efficacy against other serotypes. The predominance of G1 in this study is consistent with previous epidemiologic studies of RVGE in the United States and Finland.7 PRV efficacy was indicated by reductions in the incidence of rotavirus gastroenteritis and/or in the use of health care resources for rotavirus gastroenteritis. Efficacy has also been evaluated in a larger study in which the PRV administered at a higher potency targeted at the beginning of the vaccine's shelf life, demonstrated statistically significant efficacy against other G- serotypes.10
Based on a small subset of infants, PRV at the end of shelf-life potency was immunogenic 6 weeks after a 3-dose regimen, producing at least a threefold rise in serum anti-rotavirus IgA in >95% of PRV recipients. Some studies of naturally occurring RVGE have shown that high titers of serum anti-rotavirus IgA seem to correlate with protection against subsequent disease and/or infection.21,22 However, serologic responses to individual G-serotypes were much lower, suggesting that protective efficacy is greater than what could be inferred from serum neutralizing antibody responses.
The safety results from this study showed that the PRV at the end of shelf-life potency was generally well tolerated with respect to all adverse events (including intussusception, vomiting, diarrhea, and irritability). However, the data showed a relatively small, yet statistically significant, increase in the reporting of fevers (100.5°F, rectal equivalent) among PRV recipients compared with placebo recipients (13.4% vs 8.8%) within the 7 days of the first vaccination. This observation may be attributable to 2 reasons based on the timing of the fevers. First, the fevers reported on day 1 occurred among infants enrolled in the United States when study vaccinations were administered with routine childhood vaccinations (11.3% among PRV recipients vs 9.9% among placebo recipients) and, second, fevers reported on day 4 after vaccination coincided with the time of peak replication of the vaccine virus in the intestine. To our knowledge, this is the first and only study where such a difference in the fever rates was detected with this rotavirus vaccine. The clinical significance of this finding is probably minor because over 93% of the reported fevers were low-grade (<101.5°F, rectal equivalent) and the 5% to 8% difference observed overall and among children enrolled in the United States is less than or comparable to rates observed with other licensed childhood vaccines.23 But this is of biological interest because the replication of PRV may, in rare cases, be associated with low-grade fever.
In this study, 9 cases had vaccine-virus strain detected by EIA, but not by plaque assay, indicating that these were degraded vaccine strains; of these, 4 were within 14 days of the first dose. These cases were detected because stool samples were collected in the presence of a potential acute gastroenteritis episode. On the basis of this small number of cases, it is not possible to conclude whether or not the PRV was causally associated with the gastroenteritis symptoms.
One death from presumed SIDS was reported for an almost 5-month-old Hispanic male who was a PRV recipient. This case was not unexpected based on the background rate, gender, and age of SIDS cases reported in the literature. In 2002, the estimated rate for SIDS in the Hispanic population in the United States was 1 case per 3500 live births.24 In addition, the most common cause of death in the larger-scale study was SIDS, which occurred in 7 vaccine recipients (<0.1%) and 8 placebo recipients (<0.1%).10
This study shows that the oral pentavalent human-bovine (WC3) reassortant rotavirus vaccine was highly efficacious against severe RVGE even at the end of its shelf-life determination. Furthermore, the vaccine was immunogenic and generally well tolerated. A safe and effective vaccine targeting the major rotavirus serotypes responsible for acute gastroenteritis should have a significant impact on infant morbidity and hospitalizations.
|
ACKNOWLEDGMENTS |
|---|
We thank the US and Finnish families who enrolled their children in this study and the research nurses and coordinators who made this study possible. We especially thank Drs Tiina Korhonen (Tampere), Niklas Lindblad (Turku), and Pauli Riikonen (Pori) from Finland, whose clinics, associated with the University of Tampere, contributed to half of the study's enrollment. We also thank Donna Hyatt and Maria Petrecz for data management.
|
FOOTNOTES |
|---|
Address correspondence to Michelle G. Goveia, MD, Merck Research Laboratories, PO Box 1000, UG3CD-28, North Wales, PA 19454. E-mail: michelle_goveia{at}merck.com
Financial Disclosure: Drs Block and Vesikari are investigators and/or consultants for the sponsor. Drs Goveia, Dallas, Boslego, and Heaton and Mr Rivers, Mr Adeyi, and Mr Bauder are/were Merck employees and may hold stock and/or options in Merck & Co, Inc.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  K. S. Reisinger and S. L. Block Characteristics of an Ideal Rotavirus Vaccine Clinical Pediatrics, July 1, 2008; 47(6): 555 - 563. [Abstract] [PDF]
|
|
|
|
 |
|
 P. H. Dennehy Rotavirus Vaccines: an Overview Clin. Microbiol. Rev., January 1, 2008; 21(1): 198 - 208. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. S. Bass, D. A. Pappano, and S. G. Humiston Rotavirus Pediatr. Rev., May 1, 2007; 28(5): 183 - 191. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2007 American Academy of Pediatrics. All rights reserved. |
||
|
||
