Vaccine Research Center, University of Tampere Medical School, Tampere, Finland; 2 Wyeth Vaccine Research, Pearl River, New York; 3

Transcription

1 MAJOR ARTICLE Safety, Efficacy, and Immunogenicity of 2 Doses of Bovine-Human (UK) and Rhesus Rhesus-Human Rotavirus Reassortant Tetravalent s in Finnish Children Timo Vesikari, 1 Aino V. Karvonen, 1 Jukka Majuri, 1 Shang-Qin Zeng, 1 Xiao-Li Pang, 1 Robert Kohberger, 2 Bruce D. Forrest, 2 Yasutaka Hoshino, 3 Robert M. Chanock, 3 and Albert Z. Kapikian 3 1 Research Center, University of Tampere Medical School, Tampere, Finland; 2 Wyeth Research, Pearl River, New York; 3 Laboratory of Infectious Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland Background. Live oral rhesus rhesus-human rotavirus reassortant tetravalent (RRV-TV) vaccine was efficacious against rotavirus gastroenteritis but was withdrawn because of a rare association with intussusception. A corresponding tetravalent (types G1, G2, G3, and G4) reassortant vaccine based on bovine-human (UK) rotavirus reassortant tetravalent (BRV-TV) vaccine was developed concurrently. Methods. Before the withdrawal of RRV-TV vaccine, parallel placebo-controlled trials of BRV-TV vaccine (observer blinded) versus RRV-TV vaccine (double blinded) with a 2:1 ratio of vaccine:placebo were conducted in Finland in a total of 510 infants. Two doses of study vaccine or placebo were administered at ages 3 and 5 months. Results. The first dose of RRV-TV vaccine was followed by a significant excess rate of febrile reactions (36%), whereas the rate of fever after the administration of BRV-TV vaccine did not differ significantly from that in the placebo group. Neither vaccine induced diarrhea. A seroresponse was detected in 97% of BRV-TV vaccine recipients and 94% of RRV-TV vaccine recipients. Both vaccines were equally effective, with 68% 69% efficacy against any and 88% 100% efficacy against severe rotavirus gastroenteritis during the first epidemic season. Conclusions. BRV-TV vaccine is a promising new candidate rotavirus vaccine, with low reactogenicity and high efficacy. Two doses of BRV-TV or RRV-TV vaccine are sufficient for the induction of protection against severe rotavirus disease. Rhesus rhesus-human rotavirus reassortant tetravalent (RRV-TV) vaccine (RotaShield; Wyeth-Lederle s and Pediatrics) was licensed on 31 August 1998, and 11 Received 10 January 2006; accepted 3 March 2006; electronically published 23 June Potential conflicts of interest: T.V. has received consultancy and lecture fees from Wyeth, Merck, and GlaxoSmithKline; was an advisory board member for Wyeth; and is an advisory board member for Merck and GSK, all in relation to rotavirus vaccine. At the time of the research, R.K. was an employee of Wyeth. B.D.F. is an employee of Wyeth. Y.H., R.M.C., and A.Z.K. are employees of the US government at the National Institute of Allergy and Infectious Diseases, National Institutes of Health (NIH). The NIH has patents and a license for rhesus and rhesus-human reassortant rotavirus vaccines and patents, patent-pending applications, and licenses for bovine-human (UK) rotavirus reassortant tetravalent vaccines. A.V.K., J.M., S.-Q.Z., and X.-L.P. report no conflict of interest. Financial support: Wyeth Research. Reprints or correspondence: Prof. Timo Vesikari, University of Tampere Medical School, Biokatu 10, Tampere, Finland (timo.vesikari@uta.fi). The Journal of Infectious Diseases 2006; 194: by the Infectious Diseases Society of America. All rights reserved /2006/ $15.00 million doses of the vaccine were administered in the United States until July 1999, when its use was suspended because of a suspected association with intussusception [1]. Later the same year, the RotaShield vaccine was withdrawn by the manufacturer [2], and it has not been used since. The reintroduction of RotaShield by another manufacturer remains a possibility [3]. Before intussusception became an issue, RotaShield was regarded as safe, but its association with febrile reactions, typically 3 4 days after the administration of the first dose, was seen as a problem, at least in some populations. In Finland, administration of the first dose of vaccine at the age of 2 3 months was associated with fever (temperature, 38.0 C) in 33% of vaccinees and with a high fever (temperature, 39.0 C) in 3.4% of vaccinees [4, 5]. Such high reactogenicity correlated with low levels of preexisting rotavirus antibody in Finnish infants, compared with those in Venezuelan infants, 370 JID 2006:194 (1 August) Vesikari et al.

2 where febrile reactions were mild and were seen in only 15% of vaccinees [6, 7]. In contrast to rhesus rotavirus, bovine rotavirus candidate vaccines, such as RIT4237 and WC3, were not associated with febrile reactions [8 10]. The development of bovine rotavirus vaccine RIT4237 was discontinued, and that of the WC3 vaccine was only later resumed as a reassortant vaccine [11]. Meanwhile, the UK strain of bovine rotavirus was reassorted with human rotavirus G types 1, 2, 3, or 4, for the development of a tetravalent vaccine, at the same laboratory where the RRV-TV vaccine had originated and was evaluated in phase 2 trials [12 15]. The bovine-human (UK) rotavirus reassortant tetravalent (BRV-TV) vaccine was developed by Wyeth in collaboration with the US National Institute of Allergy and Infectious Diseases (NIAID) and is considered to be a possible future alternative for RRV-TV vaccine, because of its lower reactogenicity. The present study, which began in 1997, was conducted to evaluate and compare the reactogenicity, immunogenicity, and efficacy of 2 doses of RRV-TV and BRV-TV vaccines in Finnish infants who were 2 3 and 4 5 months old at the first and the second dose, respectively. Because of different formulations, each vaccine was compared with its own placebo, and the 2 trials were run in parallel at 2 different locations in Finland. SUBJECTS AND METHODS The study protocol was approved by the ethics committees of the Tampere University Hospital District, the City of Lahti Health Department, and the Clinical Research Committee of the NIAID. Before an infant was enrolled in the study, a parent provided written, informed consent, witnessed by a study nurse. The study was conducted between September 1997 and June Each vaccine was tested against its own placebo, randomized at a ratio of 2:1 for vaccine:placebo. The RRV-TV vaccine was tested predominantly in the city of Lahti, and the BRV-TV vaccine was tested predominantly in the city of Tampere; the 2 cities are located 130 km apart. The planned sample size was 170 for each vaccine and 85 for each placebo, for a total of 510 children. The RRV-TV vaccine and its placebo were supplied by Wyeth in identical-looking single-dose glass vials. The vaccine titer 5 5 was 4 10 pfu: 1 10 pfu for each monovalent component (rhesus-human rotavirus reassortants for types G1, G2, and G4, and rhesus rotavirus for G3), as in the RotaShield product. The placebo consisted of Eagle MEM (EMEM) with sucrose, phosphate, and glutamate (SPG). The vials were labeled with a sequential randomized subject number to which the enrolled infants were assigned. Therefore, the trial of RRV-TV vaccine versus placebo was double blinded. A buffered diluent for RRV- TV vaccine and its placebo was supplied in plastic dispettes in foil pouches, as for the RotaShield product. After reconstitution, the full content of the mixed product ( 2.5 ml) was withdrawn into the plastic dispette for oral administration. There were no feeding restrictions before or after vaccination. The BRV-TV vaccine was supplied as blocks of four 1-mL vials, each of which contained a monovalent vaccine (final passages in FRhL2 cells) with titers as follows: G1, pfu; G2, 2 10 pfu; G3, pfu; and G4, pfu. The corresponding placebo (EMEM and SPG) was also supplied as blocks of four 1-mL vials. The vaccine and placebo were prepared and supplied frozen by DynCorp. All vials (in blocks of 4) were labeled with a sequential randomized subject number to which the enrolled infants were assigned. The contents of each vial was mixed together and administered into the mouth with a syringe. Before administration, infants drank a minimum of 30 ml of soy milk buffered with 400 mg of sodium bicarbonate. The person who mixed the BRV-TV vaccine and placebo vials did not administer the vaccine; therefore, the BRV- TV vaccine versus placebo trial was observer blinded. The vaccinations were administered at dedicated vaccine research clinics of the University of Tampere in Tampere and Lahti, respectively. Only healthy infants at a gestational age of 37 weeks were enrolled. For the first dose of rotavirus vaccine or placebo, other vaccinations were not allowed within 1 week after administration of the study vaccine or placebo. For the second dose of rotavirus vaccine or placebo at 4 5 months of age, concomitant routine immunizations (with diphtheria-tetanus toxoid pertussis [DTP] vaccine and Haemophilus influenzae type b [Hib] vaccine) were allowed. The recruitment period for the first dose was from September to early December 1997 that is, before the rotavirus epidemic season. During the 7-day period after vaccination, parents recorded all events on a diary card. Rectal temperatures were measured each day at the same time or whenever the child appeared to have a fever. Parents also recorded episodes of vomiting and irritability. All stools were recorded and graded as normal or looser than normal. To enforce compliance, a phone call was made on day 5 after each dose. Venous blood samples were collected before the first immunization and 1 month (range, days) after the second immunization. Serum samples were separated, aliquoted into 2 samples, and stored frozen until analysis. One aliquot was used for the determination of serum rotavirus IgA antibodies by ELISA at the University of Tampere, as described elsewhere [16]. The other aliquot was shipped on dry ice to the Laboratory of Infectious Diseases, NIAID, for the determination of levels of rotavirus neutralizing antibodies. Levels of antibodies against each serotype of human rotaviruses G1 G4, RRV (G3), and BRV (G6) were measured under code, using a plaque-reduction neutralization assay described elsewhere [17]. Efficacy follow-up started on the fifteenth day after the second immunization. Follow-up for the first rotavirus season Bovine-Human (UK) Rotavirus JID 2006:194 (1 August) 371

3 ended on 30 June The follow-up was resumed on 1 December 1998 and continued for the second rotavirus epidemic season, until the end of June During each followup period, parents received biweekly reminder phone calls from study nurses. Whenever a child developed diarrhea, parents contacted the study nurse and started a diary card recording rectal temperature, vomiting, diarrhea, and use of oral rehydration therapy until the episode resolved; the information was used to evaluate the severity of episodes according to a numerical score described elsewhere [18]. Stool samples were obtained from all infants with gastroenteritis. The samples were kept frozen at 20 C until they were shipped to the University of Tampere Medical School for rotavirus studies. Rotavirus was primarily detected using an ELISA (Rotazyme; Dako). All positive samples were confirmed and G-typed using a reversetranscriptase polymerase chain reaction assay [19]. Postvaccination fever (rectal temperature, 38.0 C) was the primary safety parameter in the study. The analysis plan for fever specified a comparison of fever rates in infants who received placebo between the 2 sites. If not significantly different ( ), infants who received placebo would be combined. P 1.20 After each vaccination, the incidence of fever for the 7-day follow-up period was summarized, and the rates were compared using Fisher s exact test. The sample size was designed with 80% power to detect a difference in the rate of fever between BRV-TV and RRV-TV vaccine of 15% versus 30%. For immunogenicity, seroconversion by rotavirus IgA was determined as a change from negative to positive ( 20 U) [16], and the seroresponse for neutralizing antibodies against each of the G types in the reassortants vaccines was defined as a 4-fold increase. Comparison of seroresponders was done using Fisher s exact test. The primary efficacy end point was any episode of rotavirusassociated gastroenteritis. The sample size was calculated to have 80% power to detect vaccine efficacy if the true efficacy was 70%. The efficacy of each vaccine was determined on the basis of the relative risk (RR) between the vaccine group and its placebo group. efficacy was calculated as 1 RR, where the RR was the ratio of the percentage of infants in the vaccine group with the event divided by the percentage of infants in the placebo group who developed gastroenteritis. Confidence limits for RRs were calculated using exact methods as implemented in StatXact software (version 6; Cytel Software). Comparisons were done using Fisher s exact test. RESULTS In Tampere, BRV-TV vaccine was given to 172 infants, and its placebo was given to 86 infants. Dose 1 was given at a mean age of 95 days (range, days), and dose 2 was given at a mean age of 150 days (range, days). In Lahti, RRV-TV vaccine was given to 151 infants, and its placebo was given to 75 infants. Dose 1 was given at a mean age of 86 days (range, days), and dose 2 was given at a mean age of 135 days (range, days). Because recruitment in Lahti was slower than that in Tampere and the rotavirus epidemic season was approaching, the last 15 infants in the RRV-TV vaccine group were vaccinated in Tampere (10 received RRV-TV vaccine, and 5 received placebo). Data from these infants are included in the safety and immunogenicity evaluation but not in the efficacy analysis. Table 1. Reported symptoms and signs within 1 week after vaccination with bovine-human (UK) rotavirus reassortant tetravalent (BRV-TV) vaccine, rhesus rhesus-human rotavirus reassortant tetravalent (RRV-TV) vaccine, and placebo. Symptom or sign Vaccination regimen BRV-TV RRV-TV Dose 1 (n p ) (n p ) (n p ) BRV-TV vs. placebo P, Fisher s exact test RRV-TV vs. placebo BRV-TV vs. RRV-TV Fever 38.0 C !.0001!.0001 Antipyretic use !.0001!.0001 Diarrhea Vomiting Poor feeding Dose 2 (n p ) (n p ) (n p ) Fever 38.0 C Antipyretic use !.0001 Diarrhea Vomiting Poor feeding NOTE. Data are percentage of infants, unless otherwise indicated. Nos. in parentheses indicate the range of subjects with data available for each symptom. 372 JID 2006:194 (1 August) Vesikari et al.

4 Table 2. Serological responses after 2 doses of bovine-human (UK) rotavirus reassortant tetravalent (BRV-TV) vaccine, rhesus rhesus-human rotavirus reassortant tetravalent (RRV-TV) vaccine, and placebo in the perprotocol population. Assay BRV-TV (n p ) RRV-TV (n p ) (n p ) P a ELISA IgA Neutralization BRV/RRV G1 (Wa) !.0001 G2 (DS-1) !.0001 G3 (P) G4 (VA-10) Any test NOTE. Data are percentage of infants, unless otherwise indicated. Nos. in parentheses are the no. of serum samples tested for each assay/antigen. All samples were available for the ELISA IgA test, but a few were not available for the neutralization test because of exhaustion or had indeterminate results, and 1 result was excluded. a BRV-TV vs. RRV-TV, Fisher s exact test. In the safety and immunogenicity analyses, the vaccine recipients were compared with pooled placebo recipients at both sites. To justify pooling, placebo recipients in Lahti and Tampere were compared with each other for reported fever. The rates of reported fever during the first week after placebo vaccination were 12.5% and 12.2% in Lahti and Tampere, respectively (combined rate, 12.3%) (table 1); after the second placebo dose, the corresponding fever rates were 49.1% and 53.6% (combined rate, 51.6%). The striking increase in fever after the second dose most likely occurred because of the administration of DTP vaccine at the regular well-baby clinics during the same period. Safety. Rates of fever after vaccination are shown in table 1. Almost one-half of the RRV-TV vaccine recipients had a fever of 38.0 C during the week after the first vaccination, versus 12.3% of the combined placebo recipients ( P!.0001). The rate of fever in the BRV-TV vaccine recipients (16.1%) was not significantly different from that in the placebo group, whereas the difference between the RRV-TV and BRV-TV vaccinees was highly significant ( P!.0001 ). The high rate of febrile reactions after the administration of RRV-TV vaccine was substantiated by a high rate of use of antipyretics: 46.3% of the RRV-TV vaccine recipients were treated with antipyretics during the week after the first vaccination, versus 6.9% BRV-TV vaccine recipients ( P!.0001). Symptoms other than fever were not markedly different between the recipients of the RRV-TV and BRV-TV vaccines (table 1). Most notably, there was no difference in the rate of diarrhea during the week after the first vaccination, which indicates that neither vaccine caused diarrhea in the recipients. Two infants were withdrawn from the study as a consequence of a vaccine-related adverse event (both had received the RRV- TV vaccine). One child had a febrile reaction starting 3 days after the first dose of RRV-TV vaccine that lasted for 2 days, and another had unusual irritability starting 1 day after vaccination. In addition, there were a total of 14 adverse events after the administration of the first dose of study vaccine that were deemed (while coded) to be definitely related to the vaccination by the investigator. All events started on days 2 4 after dosing; were fever, irritability, or both; and lasted 1 4 days. When subject numbers were decoded, it was found that all 14 infants had received the RRV-TV vaccine. Immunogenicity. Both vaccines were immunogenic after 2 Table 3. Efficacy against rotavirus gastroenteritis of 2 doses of bovine-human (UK) rotavirus reassortant tetravalent (BRV-TV) and rhesus rhesus-human rotavirus reassortant tetravalent (RRV-TV) vaccines during the first epidemic season, in the per-protocol population. Rotavirus gastroenteritis (n p 161) BRV-TV group (n p 80) (n p 139) RRV-TV group (n p 70) Any (29 86) (28 85).005 Moderate to severe (score 7) (25 94) (66 100)!.001 Severe (score 11) (19 99) (34 100).037 NOTE. CI, confidence interval. Bovine-Human (UK) Rotavirus JID 2006:194 (1 August) 373

5 Table 4. Efficacy against rotavirus gastroenteritis of 2 doses of bovine-human (UK) rotavirus reassortant tetravalent (BRV-TV) and rhesus rhesus-human rotavirus reassortant tetravalent (RRV-TV) vaccines during the active follow-up period of 2 rotavirus seasons, in the per-protocol population. Severity of rotavirus gastroenteritis (n p 161) BRV-TV group (n p 80) (n p 139) RRV-TV group (n p 70) Any (20 80) (12 74).026 Moderate to severe (score 7) (23 93) (52 94)!.001 Severe (score 11) (36 99) (34 100).037 NOTE. CI, confidence interval. doses: the overall seroresponse rate in the per-protocol (PP) population by any test was 93.7% for RRV-TV and 96.7% for BRV-TV vaccine (table 2). ELISA IgA responses were found in 81.9% of the RRV-TV and 65.1% of the BRV-TV vaccine recipients ( P p.001). Neutralizing-antibody responses to RRV and BRV, respectively, were detected in 89.5% of the RRV-TV and 93.2% of the BRV-TV vaccine recipients. Seroresponses to human VP7 antigens were much less frequent than those to RRV or BRV. Neutralizing antibody responses to human G1 were seen in 48.7% of the BRV-TV and 23.6% of the RRV-TV vaccine recipients, respectively (P!.0001), whereas responses to G2 were observed in 35% of the RRV-TV and 12% of the BRV-TV vaccine recipients (P!.0001). Responses to G3 and G4 were not significantly different between the vaccinated groups. A seroresponse by any test was detected in 2.0% of the placebo recipients, which indicates that intercurrent natural rotavirus infections were uncommon during the vaccination period. Efficacy. There were a total of 65 episodes of acute gastroenteritis in infants who were positive by EIA for rotavirus during the entire efficacy follow-up. Of these, 48 episodes occurred during the first epidemic season, and 17 occurred during the second epidemic season. Sixty of 65 rotavirus EIA positive samples from infants with gastroenteritis were from infants who had received 2 doses of vaccine according to study protocol; they formed the PP population. Of the 60 rotavirus EIA positive samples from infants with acute gastroenteritis in the PP population 54 (90%) were of G1, 1 (2%) was of G2, 2 (3%) were of G3, and 3 (5%) were of G4. efficacy against rotavirus gastroenteritis in the PP population is presented in tables 3 and 4. efficacy in the entire (intent-to-treat [ITT]) population was similar to that in the PP population (presented below). During the first rotavirus epidemic season, the efficacy of the BRV-TV vaccine was 69% against any, 79% against moderately severe and severe, and 88% against severe rotavirus gastroenteritis. The efficacy of the RRV-TV vaccine during the first rotavirus epidemic season was 68% against any, 94% against moderately severe and severe, and 100% against severe rotavirus gastroenteritis. The difference in the efficacy of each vaccine at each end point was statistically significant (table 3). During the second rotavirus epidemic season, neither vaccine showed efficacy against any rotavirus gastroenteritis (data not shown). The point estimates for any rotavirus gastroenteritis were 0% for the BRV-TV vaccine and 16% for the RRV-TV vaccine, respectively. The point estimates for efficacy were 50% for each vaccine against moderately severe and severe rotavirus gastroenteritis. efficacy against any rotavirus gastroenteritis in the ITT population during the entire follow-up was 59% (95% confidence interval [CI], 20 79) for the BRV-TV vaccine (P p.013) and 50% (95% CI, 8 73) for the RRV-TV vaccine ( P p.030). Efficacy against rotavirus gastroenteritis with a score of 7 was 73% (95% CI, 31 91) for the BRV-TV vaccine (P p.007), and efficacy against rotavirus gastroenteritis with a score of 11 was 90% (95% CI, 35 99) for the BRV-TV vaccine Table 5. Efficacy against any gastroenteritis of 2 doses of bovine-human (UK) rotavirus reassortant tetravalent (BRV-TV) and rhesus rhesus-human rotavirus reassortant tetravalent (RRV-TV) vaccines during the entire follow-up period, in the intent-to-treat population. Gastroenteritis (n p 172) BRV-TV groups (n p 86) group (n p 151) RRV-TV groups (n p 75) Any (25 49)! ( 5 to 21).227 Moderate to severe (score 7) (9 76) (31 81).003 Severe (score 11) (35 99) (36 100).036 NOTE. CI, confidence interval. 374 JID 2006:194 (1 August) Vesikari et al.

6 ( P p.017 ) and 100% (95% CI, ) for the RRV-TV vaccine ( P p.035), respectively (data not shown). The effect of each rotavirus vaccine on all acute gastroenteritis, rotavirus or not, is presented in table 5. Only the BRV- TV vaccine showed a significant efficacy against all gastroenteritis of any severity. Both vaccines showed a significant efficacy against moderately severe and severe gastroenteritis; that of the BRV-TV vaccine was 53%, and that of the RRV-TV vaccine was 64%. DISCUSSION The present study was begun before the RRV-TV vaccine (RotaShield) was licensed and ended only weeks before the association with intussusception became known [1]. After, and as a consequence of, the withdrawal of RotaShield in October 1999, the analysis of study results was delayed. The present comparative trial of RotaShield with BRV-TV vaccine was designed as 2 parallel studies, because the 2 vaccines could not be studied in a double-blind fashion. The BRV-TV vaccine had previously been shown to be nonreactogenic but immunogenic in children [14, 15]. This was in contrast to the RRV-TV vaccine, which before licensure was already known to frequently cause febrile reactions [5]. The lower reactogenicity of the BRV-TV vaccine was confirmed in a convincing way, given that fever rates after administration of the BRV-TV vaccine were not different from those after the administration of placebo, whereas administration of the first dose of the RRV- TV vaccine was associated with a 36% rate of fever versus placebo during the first week. The low reactogenicity of the BRV-TV vaccine is in accordance with the findings of previous studies of UK bovine vaccine and is in line with the findings of a large number of previous studies of other bovine rotavirus based vaccines in humans [8 11]. Although the rhesus rotavirus [10] and RRV-TV vaccines [4, 5] were known to be reactogenic in the Finnish population, the reactogenicity of the RRV-TV vaccine was lower in the United States [20 22] and even more so in Latin America [7]. This may have been due to different prevaccination levels of rotavirus antibodies in these populations. Febrile reactions were age dependent and occurred more commonly in older infants, who have lower levels of preexisting acquired rotavirus antibodies [6]. Even in Finland, the RRV-TV vaccine did not cause febrile reactions in neonates, who have higher levels of rotavirus antibodies [23]. The immunogenicity of both vaccines, as measured by overall antibody response, was excellent and, if anything, was better for the BRV-TV vaccine. One factor contributing to the good immunogenicity of the BRV-TV vaccine might have been the large volume of buffer given before the administration of this vaccine. Both vaccines induced a strong homologous neutralizing antibody response to the animal rotavirus component (VP4) of the vaccine strain and a weaker response to human rotavirus strains G1 G4. The relatively low seroresponse rates to individual G serotypes may have been, in part, the result of technical issues, given that the neutralization test was started, due to low quantities of serum samples, from a dilution of 1: 40. However, the finding also emphasizes that mechanisms other than G type specific antibodies play an important role in vaccine-induced protection against rotavirus disease. To our knowledge, the present study is the only efficacy trial of the RRV-TV vaccine to have had a 2-dose schedule in all of the pivotal prelicensure studies in Finland, the United States, and Venezuela and in Native Americans, the vaccine was administered in 3 doses [4, 7, 20 22]. In the present study, the efficacy of 2 doses of the RRV-TV vaccine in Finnish infants was approximately the same as that in an efficacy trial of 3 doses of the same vaccine: 68% efficacy against any rotavirus gastroenteritis during the first year after vaccination in both studies [4]. As for the BRV-TV vaccine, the present results with a 2-dose schedule are not different from the recently published efficacy results of 3 doses of a pentavalent (WC3) bovine-human reassortant rotavirus vaccine, which had an efficacy against G1 G4 rotavirus gastroenteritis of any severity of 74.0% (95% confidence interval [CI], 66.8% 79.9%) and against severe G1 G4 rotavirus gastroenteritis of 98.0% (95% CI, 88.3% 100%) [24]. During the second year, the efficacy of each vaccine against mild rotavirus disease waned, but the efficacy against severe rotavirus gastroenteritis may have been sustained, and the number of cases was very small. It has been shown elsewhere [25] that, in Finnish infants, the protective efficacy of the RRV-TV vaccine against severe rotavirus gastroenteritis was maintained even during the third and fourth rotavirus epidemic seasons after vaccination. At present, it remains uncertain whether the RRV-TV vaccine will be reintroduced. It is now recognized that the risk of intussusception after administration of the RRV-TV vaccine was lower than that estimated at the time of withdrawal of RotaShield; the current estimates are between 1:10,000 and 1: 32,000, or even lower [26]. Moreover, the occurrence of intussusception was apparently dependent on age [27, 28]. To increase the safety of the RRV-TV vaccine, it would seem advisable to restrict vaccinations to infants!60 days of age, with a possible start during the neonatal period [23]. The present results indicate that a shorter schedule with 2 doses of the RRV- TV vaccine would induce an adequate level of protective efficacy. To our knowledge, this is the first demonstration of protective efficacy of the BRV-TV vaccine; it is therefore an important milestone with regard to the eventual introduction of this rotavirus vaccine. When the present study was conducted, UK bovine based vaccine was being considered by Wyeth, the manufacturer of RotaShield. More recently, licenses have been Bovine-Human (UK) Rotavirus JID 2006:194 (1 August) 375

7 granted by the US National Institutes of Health to several prospective manufacturers in the United States, Brazil, China, and India [29, 30]. The present study was a pilot study for proof of concept, because a final formulation of the BRV-TV vaccine was not tested. More important, the present study did not address safety with regard to intussusception, and a much larger study will be required for investigation of this end point. For regulatory reasons, it will be critical to establish a final composition of the future UK bovine human reassortant vaccine, possibly by adding new reassortants, such as G8 or G9 [28], to conduct the necessary safety and efficacy trials with this formulation. Acknowledgments We thank Mariam Wagner Thomas, Lashanda Long-Croal, and Jerri Ross for technical assistance. References 1. Intussusception among recipients of rotavirus vaccine United States, MMWR Morb Mortal Wkly Rep 1999; 48: Withdrawal of rotavirus vaccine recommendation. MMWR Morb Mortal Wkly Rep 1999; 48: Spiegel J. Department of Health and Human Services, National Institutes of Health. Prospective grant of exclusive license: vaccines against rotavirus disease. Federal Register 2003; 68: Joensuu J, Koskenniemi E, Pang X-L, Vesikari T. Randomised placebocontrolled trial of rhesus-human reassortant rotavirus vaccine for prevention of severe rotavirus gastroenteritis. Lancet 1997; 350: Joensuu J, Koskenniemi E, Vesikari T. Symptoms associated with rhesus-human reassortant rotavirus vaccine in infants. Pediatr Infect Dis J 1998; 17: Kapikian AZ, Hoshino Y, Flores J, et al. Alternative approaches to the development of a rotavirus vaccine. In: Holmgren J, Lindberg A, Mollby R, eds. Development of vaccines and drugs against diarrhea. 11th Nobel Conference, Stockholm Lund, Sweden: Studentlitteratur, 1986: Pérez-Schael I, Guntiñas MJ, Perez M, et al. Efficacy of the rhesus rotavirus-based quadrivalent vaccine in infants and young children in Venezuela. N Engl J Med 1997; 337: Vesikari T, Isolauri E, Delem A, d Hondt E, André FE, Zissis G. Immunogenicity and safety of live attenuated bovine rotavirus vaccine strain RIT 4237 in adults and young children. Lancet 1983; 2: Clark HF, Furukawa T, Bell LM, Offit PA, Perrella PA, Plotkin SA. Immune response of infants and children to low-passage bovine rotavirus (strain WC3). Am J Dis Child 1986; 140: Vesikari T, Kapikian AZ, Delem A, Zissis G. A comparative trial of rhesus monkey (RRV-1) and bovine (RIT 4237) oral rotavirus vaccines in young children. J Infect Dis 1986; 153: Clark HF, Offit PA, Ellis RW, et al. The development of multivalent bovine rotavirus (strain WC3) reassortant vaccine for infants. J Infect Dis 1996; 174(Suppl 1):S Midthun K, Greenberg HB, Hoshino Y, Kapikian AZ, Wyatt RG, Chanock RM. Reassortant rotaviruses as potential live rotavirus vaccine candidates. J Virol 1985; 53: Midthun K, Hoshino Y, Kapikian AZ, Chanock RM. Single gene substitution rotavirus reassortants containing the major neutralization protein (VP7) of human rotavirus serotype 4. J Clin Microbiol 1986; 24: Clements-Mann ML, Makhene MK, Mrukowicz J, et al. Safety and immunogenicity of live attenuated human-bovine (UK) reassortant rotavirus vaccines with VP7-specificity for serotypes 1, 2, 3 or 4 in adults, children and infants. 1999; 17: Clements-Mann ML, Dudas R, Hoshino Y, et al. Safety and immunogenicity of live attenuated quadrivalent human-bovine (UK) reassortant rotavirus vaccine administered with childhood vaccines to infants. 2001; 19: Isolauri E, Joensuu J, Suomalainen H, Luomala M, Vesikari T. Improved immunogenicity of oral D RRV reassortant rotavirus vaccine by Lactobacillus casei GG. 1995; 13: Wyatt RG, Greenberg HB, James WD, et al. Definition of human rotavirus serotypes of plaque reduction assay. Infect Immun 1982; 37: Ruuska T, Vesikari T. Rotavirus disease in Finnish children: use of numerical scores for clinical severity of diarrhoeal episodes. Scand J Infect Dis 1990; 22: Pang XL, Joensuu J, Hoshino Y, Kapikian AZ, Vesikari T. Rotaviruses detected by reverse transcription polymerase chain reaction in acute gastroenteritis during a trial of rhesus-human reassortant rotavirus tetravalent vaccine: implications for vaccine efficacy analysis. J Clin Virol 1999; 13: Bernstein DI, Glass RI, Rodgers G, Davidson BL, Sack DA. Evaluation of rhesus rotavirus monovalent and tetravalent reassortant vaccines in US children. US Rotavirus Efficacy Group. JAMA 1995; 273: Rennels MB, Glass RI, Dennehy PH, et al. Safety and efficacy of highdose rhesus-human reassortant rotavirus vaccines report of the National Multicenter Trial. United States Rotavirus Efficacy Group. Pediatrics 1996; 97: Santosham M, Moulton LH, Reid R, et al. Efficacy and safety of highdose rhesus-human reassortant rotavirus vaccine in Native American populations. J Pediatr 1997; 131: Vesikari T, Karvonen A, Forrest BD, Hoshino Y, Chanock RM, Kapikian AZ. Neonatal administration of rhesus rotavirus tetravalent vaccine. Pediatr Infect Dis J 2006; 25: Vesikari T, Matson DO, Dennehy P, et al. Safety and efficacy of a pentavalent human-bovine (WC3) reassortant rotavirus vaccine. N Engl J Med 2006; 354: Joensuu J, Koskenniemi E, Vesikari T. Prolonged efficacy of rhesus-human reassortant rotavirus vaccine. Pediatr Infect Dis J 1998; 17: Murphy BR, Morens DM, Simonsen L, et al. Reappraisal of the association of intussusception with the licensed live rotavirus vaccine challenges initial conclusions. J Infect Dis 2003; 187: Simonsen L, Viboud C, Elixhausen A, Taylor RJ, Kapikian AZ. More on RotaShield and intussusception: the role of age at vaccination. J Infect Dis 2005; 192(Suppl):S Kapikian AZ, Simonsen L, Vesikari T, et al. A hexavalent human rotavirus bovine rotavirus (UK) reassortant vaccine designed for use in developing countries and delivered in a schedule with the potential to eliminate the risk of intussusception. J Infect Dis 2005; 192(Suppl): S Rohrbaugh ML. Department of Health and Human Services, National Institutes of Health. Prospective grant of exclusive license: humanbovine reassortant rotavirus vaccine. Federal Register 2004; 69: Ferguson SM. Department of Health and Human Services, National Institutes of Health. Prospective grant of partially exclusive licences: human-bovine reassortant vaccine. Federal Register 2004; 69: JID 2006:194 (1 August) Vesikari et al.