A Bovine Parainfluenza Virus Type 3 Vaccine Is Safe and Immunogenic in Early Infancy

Transcription

1 MAJOR ARTICLE A Bovine Parainfluenza Virus Type 3 Vaccine Is Safe and Immunogenic in Early Infancy David P. Greenberg, 1,a Robert E. Walker, 3 Min-Shi Lee, 3 Keith S. Reisinger, 2 Joel I. Ward, 4 Ram Yogev, Mark M. Blatter, 2 Sylvia H. Yeh, 4 Ruth A. Karron, Chithra Sangli, 3,a Lane Eubank, 3 Kathleen L. Coelingh, 3 Julie M. Cordova, 3 Marilyn J. August, 3 Harshvardhan B. Mehta, 3 Wendy Chen, 3 and Paul M. Mendelman 3 1 University of Pittsburgh School of Medicine, Children s Hospital of Pittsburgh, and 2 Primary Physicians Research, Pittsburgh, Pennsylvania; 3 MedImmune Vaccines, Inc., Mountain View, and 4 UCLA Center for Vaccine Research, Research and Education Institute at Harbor-UCLA, Torrance, California; Children s Memorial Hospital, Chicago, Illinois; Center for Immunization Research, Johns Hopkins University, Baltimore, Maryland Background. A phase 2 trial was conducted to assess in young infants the safety, tolerability, infectivity, and immunogenicity of multiple doses of an intranasal vaccine using bovine parainfluenza virus type 3 (bpiv3). Methods. One hundred ninety-two healthy 2-month-old infants were randomized 1:1:1 to receive 1 10 median tissue culture infective dose (TCID 0 ) bpiv3 vaccine, 1 10 TCID 0 bpiv3 vaccine, or placebo at 2, 4,, and 12 1 months of age. Safety information was collected by use of diary sheets and telephone interviews. Nasal wash and serum specimens were collected for assessment of infectivity and immunogenicity. Results. The safety profiles of both dosages of bpiv3 were similar to that of placebo, with the exception of fever with temperature of 38.1 C after dose 2 only, occurring in 34% of the 1 10 TCID 0 group, 3% of the 1 10 TCID 0 group, and 12% of the placebo group ( P!.01). No vaccine-related serious adverse events were reported. The cumulative vaccine infectivity (isolation of bpiv3 and/or bpiv3 seroconversion) after dose 3 was similar in the 2 vaccine groups (87% in the 1 10 TCID 0 group and 77% in the 1 10 TCID 0 group) ( P p.4). Seroconversion rates after dose 3, assessed by means of hemagglutination inhibition assay, after adjustment for decrease in maternal antibody titers, were 7% in the 1 10 TCID 0 group, 7% in the 1 10 TCID 0 group, and 12% in the placebo group ( P!.01 ). of bpiv3 was common after dose 1, dose 2, or dose 3, but only 1 of 1 participants in the vaccine groups had bpiv3 isolated after dose 4. Conclusions. Multiple doses of bpiv3 vaccine were well tolerated and immunogenic in young infants. Human parainfluenza virus type 3 (hpiv3) is an important respiratory pathogen in young children and accounts for 11% of pediatric hospitalizations for acute respiratory tract illnesses [1 2]. hpiv3 is second only Received 4 June 2004; accepted 13 October 2004; electronically published 22 February 200. Presented in part: Pediatric Academic Societies annual meeting, San Francisco, 1 4 May 1999; 4th Annual Conference on Vaccine Research, Arlington, Virginia, 23 2 April 2001 (abstract 38); 1st International Neonatal Vaccination Workshop, McLean, Virginia, 2 4 March 2004 (abstract 0104). Potential conflicts of interest: D.P.G., K.S.R., J.I.W., R.Y., M.M.B., and S.H.Y. were investigators in this clinical study; R.A.K. was a consultant to MedImmune Vaccines, Inc., during the study; R.E.W., M.-S.L., C.S., L.E., K.L.C., J.M.C., M.J.A., H.B.M., W.C., and P.M.M. were employees of MedImmune Vaccines, Inc., during the study; MedImmune Vaccines, Inc. is developing a PIV3 vaccine. Financial support: MedImmune Vaccines, Inc. a Present affiliation: Sanofi Pasteur, Inc., Swiftwater, Pennsylvania (D.P.G.); Genomic Health, Inc., Redwood City, California (C.S.). Reprints or correspondence: Dr. Min-Shi Lee, 297 N. Bernardo Ave., Mountain View, CA (leem@medimmune.com or mslee007@us.sina.com). The Journal of Infectious Diseases 200; 191: by the Infectious Diseases Society of America. All rights reserved /200/ $1.00 to respiratory syncytial virus (RSV) as a leading cause of bronchiolitis and pneumonia in young infants [2 4]. Because of the burden of disease associated with hpiv3 in young infants, developing a safe and effective vaccine is an important public health priority. Bovine parainfluenza virus type 3 (bpiv3) has been in clinical development for use in a live intranasal vaccine to protect against hpiv3 disease. The rationale for this approach is that (1) replication of bpiv3 is attenuated in nonhuman primates, presumably because of host-range restriction; (2) bpiv3 and hpiv3 share neutralization epitopes []; (3) vaccination of primates with bpiv3 induces cross-protective immunity against challenge with hpiv3 [ 7]; and (4) intranasally administered viral vaccines may offer the advantage of inducing local mucosal immunity, which may correlate with protection against hpiv3 disease. Furthermore, current advances in genetic engineering techniques have presented the possibility of using a suitably atten- Downloaded from at Pennsylvania State University on May 8, JID 200:191 (1 April) Greenberg et al.

2 Table 1. Schedule for 10 visits in a clinical trial of bovine parainfluenza virus type 3 vaccine. Visit Age, months Study day Vaccination X X X X Serum collection X X X X X Nasal wash a X X X X Note. Routine pediatric vaccines were administered at 2, 4,, and 12 1 months of age. X, procedure performed. a Performed at 3 of the 4 investigative sites. uated bpiv3 vaccine as a vector to deliver hpiv3 antigens and/ or RSV antigens to protect against infection with hpiv3 and/ or RSV [8, 9]. In phase 1 studies, 1 or 2 doses of bpiv3, administered by intranasal drops, were demonstrated to be safe and immunogenic in limited numbers of adults, children, and infants [10 12]. The present study was designed to assess the safety, tolerability, infectivity, and immunogenicity of multiple doses of bpiv3 vaccine, administered by intranasal spray, to young infants and to select an optimal dose and schedule for a primary immunization series. SUBJECTS, MATERIALS, AND METHODS Study population. Healthy infants 2 months of age were eligible to participate. Potential subjects were excluded from participation in the event of (1) immunodeficiency or immunosuppressive therapy of the child or a household member, (2) febrile or upper respiratory tract illness 3 days before enrollment into the study, (3) use of intranasal medications 1 week before enrollment into the study, or (4) previous receipt of blood products 2 months before enrollment into the study or expected receipt of blood products at any time during the study period. Study design. The study was conducted at 4 clinical study centers in the United States. Enrollment began in the fall of 1997, and participation was completed in the spring of The protocol was approved by accredited institutional review boards responsible for each center, and signed informed consent was obtained from each child s parent or guardian. The human-experimentation guidelines of the US Department of Health and Human Services were followed in the conduct of the clinical research. Subjects were randomly assigned in equal proportions, by use of computer-generated numbers, to 1 of 3 groups receiving either intranasal placebo (tissue culture medium) or 1 10 TCID 0 or 1 10 TCID 0 of the bpiv3 Kan- sas/12/84 strain grown in fetal rhesus monkey lung cells. MedImmune Vaccines, Inc. (Mountain View, CA) acquired the rights to the bpiv3 vaccine from the National Institutes of Health (NIH) and used the same vaccine materials that were previously tested in the NIH phase 1 and phase 2 studies [10 12]; the vaccine was retested for potency and packaged into nasal sprayers. Each dose was 0.2 ml (0.1 ml/nostril) and was administered by intranasal spray by use of the Becton Dickinson Accuspray that is identical to that used to administer the live attenuated influenza vaccine (FluMist; MedImmune Vaccines) [13]. A total of 4 doses of vaccine or placebo, depending on group assignment, were administered at 2, 4,, and 12 1 months of age concurrently with routine pediatric vaccines, following the American Academy of Pediatrics recommendations (i.e., diphtheria, tetanus, acellular pertussis [DTaP], Haemophilus influenzae type b conjugate [Hib], inactivated or oral polio virus [IPV or OPV], hepatitis B, measles, mumps, rubella, and varicella vaccines) (table 1). Parents/guardians and study staff were blinded to the administered treatment. Parents/guardians recorded potential adverse events on diary cards for 14 days after each dose of the vaccine and were contacted by telephone 14 and 30 days after each dose of the vaccine, to capture additional adverse events. Five serum specimens were collected from the participants: at baseline immediately before dose 1, at 2 months of age; before and 1 month after dose 3, at months of age; and before and 1 month after dose 4 (booster), at 12 1 months of age. At 3 of the 4 investigative sites, a single nasal wash specimen, for isolation of bpiv3, was collected from each subject 7 days after each dose. Virus culture specimens were collected for outpatient illnesses during the safety assessment periods (28 days after each dose), and these specimens were not collected to assess vaccine efficacy. Virological assays. Undiluted nasal wash specimens were stored frozen at 70 C until analyzed. of bpiv3 was performed by use of Madin-Darby bovine kidney cells. Viral isolates were identified as bpiv3 (vaccine) or hpiv3 (wild-type) by use of indirect immunofluorescence assay with specific monoclonal antibodies [10 12]. Serological assays. Hemagglutination inhibition (HI) antibodies against bpiv3 and hpiv3 antigens were measured by use of 0.% guinea pig red blood cells and 4 hemagglutina- Downloaded from at Pennsylvania State University on May 8, 201 PIV3 Vaccine in Early Infancy JID 200:191 (1 April) 1117

3 Table 2. Subjects with solicited adverse events within 14 days after vaccination, by dose and treatment group. Adverse event 1 10 TCID TCID 0 Placebo After dose 1 n p 4 n p 1 n p Temperature 38.1 C 22 (13 34) 20 (11 32) 14 ( 24) Temperature 39.0 C 0 (0 ) (1 14) 3 (0 11) Cough 33 (22 4) 31 (20 44) 27 (17 40) Runny nose/nasal congestion 48 (3 1) 43 (30 ) 3 (40 ) Irritability 39 (27 2) 4 (33 9) 44 (32 7) Vomiting 9 (4 19) 11 ( 22) 12 ( 22) Decreased activity 9 (4 19) 18 (9 30) 12 ( 22) After dose 2 n p 9 n p n p 8 Temperature 38.1 C 34 (22 47) a 3 (22 49) a 12 ( 23) a Temperature 39.0 C 10 (4 21) 9 (3 20) 3 (0 12) Cough 2 (1 38) b 27 (1 41) b 12 ( 23) b Runny nose/nasal congestion 29 (18 42) 27 (1 41) 29 (18 43) Irritability 31 (19 44) 38 (2 2) 31 (20 4) Vomiting 7 (2 1) 7 (2 18) 12 ( 23) Decreased activity 7 (2 1) 13 ( 24) 10 (4 21) After dose 3 n p n p 1 n p Temperature 38.1 C 18 (9 30) c 29 (17 44) c 41 (28 ) c Temperature 39.0 C 2 (0 10) 10 (3 21) 7 (2 17) Cough 20 (10 32) 22 (11 3) 21 (12 34) Runny nose/nasal congestion 39 (2 3) 39 (2 4) 34 (22 48) Irritability 32 (20 4) 39 (2 4) 41 (28 ) Vomiting 13 ( 24) 12 (4 24) (1 1) Decreased activity (1 1) 10 (3 21) 9 (3 20) After dose 4 n p 3 n p 4 n p 49 Temperature 38.1 C 30 (18 44) d 28 (1 43) d (40 9) d Temperature 39.0 C 1 (7 28) 11 (4 24) 24 (13 39) Cough 38 (2 2) 30 (18 4) 33 (20 48) Runny nose/nasal congestion 1 (37 ) 39 (2 ) 39 (2 4) Irritability 38 (2 2) 24 (13 39) 3 (22 0) Vomiting 1 (7 28) 11 (4 24) 12 ( 2) Decreased activity 11 (4 23) 13 ( 2) 24 (13 39) NOTE. Data are % (9% confidence interval). a By Fisher s exact test: P p.00, for 3 groups overall; P p.008, for 1 10 TCID 0 vs. placebo; P p.007, for 1 10 TCID 0 vs. placebo. b By Fisher s exact test: P p.084 for 3 groups overall; P p.097, for 1 10 TCID 0 vs. placebo; P p.07, for 1 10 TCID 0 vs. placebo. c By Fisher s exact test:, for 3 groups overall;, for P p.02 P p TCID vs. placebo; 0 P p.231, for 1 10 TCID 0 vs. placebo. d By Fisher s exact test:, for 3 groups overall;, for P p.011 P p TCID vs. placebo; 0 P p.012, for 1 10 TCID 0 vs. placebo. ternal origin, were used as the baseline to define hpiv3 seroconversion at, 7, 12 1, and 13 1 months of age, after adjustment of hpiv3 HI antibody titers for a 1-day biological half-life of maternal hpiv3 HI antibody titers [1, 1]. bpiv3 HI antibodies were also detected at 2 months of age, because of cross-reactivity between maternal hpiv3 antibodies and bpiv3 antigens. Therefore, bpiv3 seroconversion was cal- Downloaded from at Pennsylvania State University on May 8, 201 tion units of hpiv3/washington/197 strain or bpiv3/kansas/ 12/84 strain viruses. The antibody titer was defined as the reciprocal of the highest serum dilution that completely inhibited hemagglutination, according to the standard procedure [14]. The starting serum dilution was 1:4, and the cutoff level of seropositivity was set at 4 [1, 1]. Titers of hpiv3 HI antibodies found at 2 months of age, considered to be of ma JID 200:191 (1 April) Greenberg et al.

4 Table 3. Correlation between bovine parainfluenza virus type 3 (bpiv3) hemagglutination inhibition (HI) antibody titers at 2 months of age and occurrence of adverse events (fever with temperature 38.1 C and/ or cough) and isolation of bpiv3 after dose 1 and dose 2. Table 4. either. Time of measurement Adverse events Virus shedding 1 10 TCID TCID TCID TCID 0 bpiv3 HI antibody titers Dose 1 Dose 2 Dose 1 Dose 2 Dose 1 Dose 2 Dose 1 Dose 2 No. of participants/total no. of participants (%)!4 9/18 (0) 10/18 () 9/21 (43) 9/17 (3) 3/8 (38) 3/7 (43) /10 (0) 4/9 (44) 4 /19 (32) /18 (28) 9/18 (0) 7/1 (47) 7/10 (70) /11(4) /9 () 1/8 (13) 8 /12 (0) /11 () /13 (38) 8/13 (2) 1/7 (14) 0/ (0) /8 (9) 2/7 (29) 1 2/ (40) 2/ (40) 2/ (40) 2/ (40) 1/ (17) 2/ (40) 1/2 (0) 2/4 (0) Significance a Mantel-Haenszel x 2 test for trend. culated after adjustment for maternal bpiv3 HI antibody titers by use of the same adjustment method. bpiv3 seroconversion was defined as a 4-fold increase in bpiv3 HI antibody titers after adjustment for maternal bpiv3 HI antibody titers. Statistical analysis. Rates of adverse events were compared overall for the 3 treatment groups by use of 2-tailed Fisher s exact test. If statistical significance ( P.0) or marginal significance was observed, pairwise comparisons between the groups were conducted. Seroconversion rates and rates of isolation of bpiv3 between treatment groups were also compared by use of 2-tailed Fisher s exact test. Exact 9% confidence intervals were obtained for these rates. hpiv3 and bpiv3 seroconversion rates within the same treatment group were compared by use of McNemar s test. A dose-response relationship was tested by use of the Mantel-Haenszel x 2 test for trend. Statistical analyses were performed by use of SAS software (ver- No. and percentage of subjects with isolation of bovine parainfluenza virus type 3 (bpiv3), bpiv3 seroconversion, or of bpiv3 P a 1 10 TCID TCID 0 bpiv3 seroconversion a sion.12; SAS Inc.) and StatXact software (version 3.1; Cytel Software). RESULTS Demographic characteristics. A total of 192 participants 4 in the 1 10 TCID 0 group, 2 in the 1 10 TCID 0 group, and in the placebo group were enrolled. The 3 treatment groups were similar with respect to mean age at dose 1 of the vaccine (9.1, 9.2, and 9.4 weeks, respectively), sex (%, 42%, and 2% male, respectively), and race/ethnicity (48%, 39%, and 44% Hispanic, respectively; 27%, 32%, and 3% white, respectively). Safety. Incidence rates of solicited adverse events after each dose of the vaccine are shown in table 2. After dose 1, the most frequently reported solicited adverse events in all 3 groups were of bpiv3 or bpiv3 seroconversion b of bpiv3 bpiv3 seroconversion a of bpiv3 or bpiv3 seroconversion b After dose 1 12/32 (38) NA c NA c 17/30 (7) NA c NA c After dose 2 11/30 (37) NA c NA c 10/29 (34) NA c NA c After dose 3 7/28 (2) 10/1 (19) 9/27 (33) 7/28 (2) 7/47 (1) 7/2 (28) Cumulative (doses 1 3) 19/24 (79) d 32/48 (7) e 20/23 (87) 19/24 (79) d 27/47 (7) e 17/22 (77) After dose 4 0/27 (0) 7/3 (13) 2/27 (7) 1/24 (4) /39 (13) 4/19 (21) Downloaded from at Pennsylvania State University on May 8, 201 NOTE. Data are no. of participants/total no. of participants (%). NA, not available. a Defined as a 4-fold increase in bpiv3 hemagglutination inhibition antibody titers between vaccinations, after adjustment for decrease in maternal antibody titers. b Only participants who had available nasal washes and serum specimens are included. c NA because serum specimens were not collected after dose 1 and before dose 2. d Only participants with all 3 nasal wash specimens are included. e Calculated by comparing bpiv3 HI antibody titers before dose 1 and after dose 3, after adjustment for decrease in maternal antibody titers. PIV3 Vaccine in Early Infancy JID 200:191 (1 April) 1119

5 Table. Correlation between preexisting bovine parainfluenza virus type 3 (bpiv3) hemagglutination inhibition (HI) antibody titers and bpiv3 seroconversion after dose 3. bpiv3 HI antibody titer before dose 3: months of age bpiv3 seroconversion after dose TCID TCID 0 No. of participants/ total no. of participants (%)!4 9/21 (43) /14 (3) 4 1/20 () 2/18 (11) 8 0/9 (0) 0/8 (0) 1 0/1 (0) 0/7 (0) P a Significance a Mantel-Haenszel x 2 test for trend. runny nose/nasal congestion, irritability, and cough. In general, these events occurred at comparable rates in all groups after all 4 doses except after dose 2, when a significantly higher incidence of fever with temperature 38.1 C was found in the vaccine groups compared with the placebo group ( P!.01, for each vaccine group compared with the placebo group) (table 2). The elevated temperatures occurred primarily on day 0 or day 1 after dose 2. There were no significant differences between the treatment groups in incidence of fever at higher temperature thresholds ( 39.0 C) after dose 2 (table 2). Interestingly, the placebo group had a significantly higher incidence of fever 14 days after dose 3 than the 1 10 TCID 0 group (41% vs. 18%) ( P p.012, by Fisher s exact test), and the placebo group had a significantly higher incidence of fever 14 days after dose 4 than both vaccine groups (% in the placebo group vs. 30% in the 1 10 TCID 0 group and 28% in the 1 10 TCID 0 group) ( P!.0, by Fisher s exact test) (table 2). After dose 2, cough occurred in 2% of the 1 10 TCID 0 group and in 27% of the 1 10 TCID 0 group, compared with 12% of the placebo group, but these differences were not statistically significant (P p.097 and P p.07, respectively, compared with the placebo group; P p.084, for 3 groups overall, by Fisher s exact test) (table 2). The cough episodes were evenly distributed over the 14 days. Rates of cough were comparable in the vaccine groups and the placebo group after dose 1, dose 3, and dose 4. It is possible that high maternal bpiv3 HI antibody titers protected 2-month-old infants against fever with temperature 38.1 C and/or cough after dose 1 and that, by 4 months of age, the titers decreased sufficiently for the rates of these adverse events to increase after dose 2. To evaluate this possibility, we categorized bpiv3 HI antibody titers at 2 months of age into 4 groups (!4, 4, 8, and 1). No correlation between bpiv3 HI antibody titers at 2 months of age and incidence of fever with temperature 38.1 C and/or cough after dose 1 or dose 2 was observed in either vaccine group (table 3), which supports the hypothesis that the higher incidence of fever or cough after dose 2 was not related to the vaccine. Eight serious adverse events, all of which resulted in hospitalizations, were reported, and 0 were classified as being vaccine-related. These included 4 events in 4 children in the 1 10 TCID 0 group (parainfluenza type 1 culture-positive croup 8 days after dose 1, Streptococcus pneumoniae bacterial meningitis 30 days after dose 1, otitis media 12 days after dose 2, and elective surgery for correction of syndactyly 20 days after dose 4), 2 events in 2 children in the 1 10 TCID 0 group (retropharyngeal abscess 11 days after dose 2 and acute otitis media and rotavirus culture-positive gastroenteritis in 1 child 29 days after dose 2), and 2 events in 1 child in the placebo group (bronchiolitis due to RSV at 38 days after dose 1 and an abdominal mass subsequently diagnosed as neuroblastoma presenting at 27 days after dose 2). Infectivity. The infectivity of the bpiv3 vaccine was assessed by measurement of the proportion of study participants who either had bpiv3 isolated in nasal wash specimens or had a 4-fold increase in bpiv3 HI antibody titers (table 4). This analysis was restricted to participants at the 3 sites at which isolation of bpiv3 was assessed. After dose 1, at 2 months of age, 38% of the 1 10 TCID 0 group and 7% of the 1 10 TCID 0 group had bpiv3 isolated ( P p.20, by Fisher s exact test). After dose 2, at 4 months of age, 37% of the 1 10 TCID 0 group and 34% of the 1 10 TCID 0 group had bpiv3 isolated. After dose 3, a similar proportion of participants in each vaccine group (2%) had bpiv3 isolated. After dose 4, only 1 participant in the 1 10 TCID 0 group had bpiv3 isolated. After dose 3, 0 participants in the placebo group had bpiv3 isolated, and 1 participant had hpiv3 isolated at day. The cumulative proportion of those in the vaccine groups who had bpiv3 isolated after dose 1, dose 2, or dose 3 was the same (79%). After dose 3, bpiv3 seroconversion rates in the vaccine groups were similar (7% in the 1 10 TCID 0 group and 7% in the 1 10 TCID 0 group) ( P p.40, by Fisher s exact test). When rates of isolation of bpiv3 or of seroconversion were combined, after dose 3, the cumulative infectivity rate was 87% in the 1 10 TCID 0 group and 77% in the 1 10 TCID 0 group ( P p.4, by Fisher s exact test). After dose 4, only 13% of participants in the vaccine groups developed increases in bpiv3 HI antibody titers. To evaluate the correlation between passively acquired maternal bpiv3 HI antibodies and isolation of bpiv3 after dose 1 or dose 2, we categorized bpiv3 HI antibody titers at 2 months of age into 4 groups (!4, 4, 8, and 1). No correlation between bpiv3 HI antibody titers at 2 months of age and isolation of bpiv3 after dose 1 or dose 2 was observed in either vaccine group (table 3). The correlation between preexisting Downloaded from at Pennsylvania State University on May 8, JID 200:191 (1 April) Greenberg et al.

6 bpiv3 HI antibody titers and development of bpiv3 seroconversion after dose 1 or dose 2 cannot be evaluated, because serum specimens were not collected between dose 1 and dose 2. However, we were able to assess the impact of bpiv3 HI antibody titers on bpiv3 seroconversion after dose 3. Participants in the vaccine groups who had lower preexisting bpiv3 HI antibody titers had significantly higher bpiv3 seroconversion rates than participants in the vaccine groups who had higher preexisting HI bpiv3 antibody titers ( P p.002, for the 1 10 TCID 0 group; P p.013, for the 1 10 TCID 0 group, by Mantel-Haenszel x 2 test for trend) (table ). Immunogenicity. bpiv3 and hpiv3 seroconversion rates are shown in table. At all 4 time points, bpiv3 seroconversion rates were significantly higher than hpiv3 seroconversion rates in the vaccine groups, indicating that an antibody response to the bpiv3 vaccine was more likely to be detected by the bpiv3 HI assay than by the hpiv3 HI assay. bpiv3 seroconversion rates increased steadily from 4% before dose 3, at months of age, to 8% after dose 4, at 13 1 months of age, in the 1 10 TCID 0 group, and from 48% to 100%, respectively, in the 1 10 TCID 0 group; these rates in the vaccine groups were significantly higher than those in the placebo group (9% before dose 3, at months of age, and 0% after dose 4, at 13 1 months of age) ( P!.0, by Fisher s exact test). No vaccine was given between 7 months of age and 12 1 months of age, but hpiv3 seroconversion rates during the interval increased from 1% to 47% in the placebo group, from 2% to Table. Human parainfluenza virus type 3 (hpiv3) and bovine parainfluenza virus type 3 (bpiv3) seroconversion rates, compared with baseline titers. Seroconversion 1% in the 1 10 TCID 0 group, and from 21% to 4% in the 1 10 TCID 0 group, indicating that exposure to hpiv3 was likely during the interval (table ). DISCUSSION The present study showed that the bpiv3 vaccine was well tolerated and immunogenic in young infants. In a previous phase 1 study of 19 infants 2 months of age, the incidence of fever was not significantly higher in the 1 10 TCID 0 group after dose 1, compared with that in the placebo group [11]. Although our data after dose 1 are consistent with those in the previous study [11], a higher incidence of low-grade fever after dose 2 in the vaccine groups was observed in the present study. This higher incidence of low-grade fever may be related to the combined effects of concurrent vaccination with multiple other vaccines (e.g., DTaP, Hib, IPV, or OPV) or may have occurred by chance. Similarly, the higher incidence of fever in the placebo group after dose 3 and dose 4 may be due to chance alone. Larger sample sizes will be required in future studies, to clarify any cause-effect relationship. After dose 3, 82% (37/4) of those in the vaccine groups were infected with bpiv3, as measured by rates of isolation of bpiv3 or bpiv3 seroconversion, whereas, after dose 4, only low proportions of those in the vaccine groups had increased bpiv3 HI antibody titers (12/92; 13%) or shed virus (1/1; 2%) (table 4, data combined from the 1 10 TCID 0 and 1 10 TCID TCID TCID 0 Placebo Before dose 3: months of age n p 4 n p 1 n p 4 hpiv3 13 a 10 a 11 bpiv3 48 a,b 4 a,b 9 b After dose 3: 7 months of age n p 47 n p 48 n p 0 hpiv3 21 a 2 a 1 bpiv3 7 a,b 7 a,b 12 b Downloaded from at Pennsylvania State University on May 8, 201 Before dose 4: 12 1 months of age n p 39 n p 49 n p 43 hpiv3 4 a 1 a 47 bpiv3 82 a 8 a 47 After dose 4: 13 1 months of age n p 40 n p 49 n p 44 hpiv3 0 a 9 a 0 bpiv3 100 a,b 8 a,b 0 b NOTE. Data are %. The details for defining bpiv3 and hpiv3 seroconversion are described in the Subjects, Materials, and Methods section. a P!.0, for hpiv3 seroconversion vs. bpiv3 seroconversion in the same treatment group at each time point. b P!.01, for bpiv3 seroconversion in the placebo group vs. the 1 10 TCID 0 group or 1 10 TCID 0 group. PIV3 Vaccine in Early Infancy JID 200:191 (1 April) 1121

7 groups). These data support the hypothesis that 3 primary doses likely provide sufficient immunity against hpiv3. Although data on seroconversion was combined with data on isolation of bpiv3 for the purpose of detecting vaccine infectivity in the present study, these rates may still underestimate the vaccine s infectivity, because viral cultures were obtained only once after each dose. The lower rates of isolation of bpiv3 after dose 1 in the present study (38% 7% in the 2 vaccinated groups) is in contrast to those in a previous study of this vaccine, in which 11 of 12 vaccinees (92%) (2 months of age; mean age, 4.3 months) had bpiv3 isolated after dose 1 [11]. This difference may be explained by less frequent collection of nasal wash specimens (only once in the week after vaccination in the present study, compared with 4 specimens obtained 11 days after each dose of vaccine in the previous study). Previous studies and the present study have shown bpiv3 to be safe and immunogenic in young infants [10 12]. Recent in vitro and animal studies have confirmed the potential of bpiv3 vaccine to serve as a vector to express antigens of hpiv3 and RSV [8, 9]. Future development of the bpiv3 vaccine as a vector for hpiv3, RSV, and other respiratory virus antigens in humans is warranted. Acknowledgment We thank the study coordinators, referring pediatricians, and parents of participating infants. Karen Smail, Janet Mugger, Jill Schwalb, Nancy Lodell, Julie Iwashita, Iksung Cho, Stephen Chan, Denise Dawson, Tim Kang, Katie Komaroff, Sharon Mathie, Sandra Holmes, and Mengshu Zhao also contributed to this study. References 1. US Institute of Medicine. New vaccine development: establishing priorities: diseases of importance in the United States. Washington, DC: National Academy Press, Glezen WP, Denny FW. Epidemiology of acute lower respiratory disease in children. N Engl J Med 1973; 288: Reed G, Jewett PH, Thompton J, Tollefson S, Wright PF. Epidemiology and clinical impact of parainfluenza virus infections in otherwise healthy infants and young children! years old. J Infect Dis 1997; 17: Glezen WP, Denny FW. Parainfluenza viruses. In: Evans AS, Kaslow RA, eds. Viral infections of humans: epidemiology and control. 4th ed. New York: Plenum, 1997:1 7.. Coelingh KJ, Winter CC, Murphy BR, et al. Conserved epitopes on the hemagglutinin-neuraminidase proteins of human and bovine parainfluenza type 3 viruses: nucleoide sequence analysis of variants selected with monoclonal antibodies. J Virol 198; 0:90.. van Wyke Coelingh KL, Winter CC, Tierney EL, London WT, Murphy BR. Attenuation of bovine parainfluenza virus type 3 in nonhuman primates and its ability to confer immunity to human parainfluenza virus type 3 challenge. J Infect Dis 1988; 17: Pennathur S, Haller AA, MacPhail M, et al. Evaluation of attenuation, immunogenicity and efficacy of a bovine parainfluenza virus type 3 (PIV-3) vaccine and a recombinant chimeric bovine/human PIV-3 vaccine vector in rhesus monkeys. J Gen Virol 2003; 84: Haller AA, Miller T, Mitiku M, Coelingh K. Expression of the surface glycoproteins of human parainfluenza virus type 3 by bovine parainfluenza virus type 3, a novel attenuated virus vaccine vector. J Virol 2000; 74: Schmidt AC, McAuliffe JM, Murphy BR, Collins PL. Recombinant bovine/human parainfluenza virus type 3 (B/HPIV3) expressing the respiratory syncytial virus (RSV) G and F proteins can be used to achieve simultaneous mucosal immunization against RSV and HPIV3. J Virol 2001; 7: Karron RA, Wright PF, Hall SL, et al. A live attenuated bovine parainfluenza virus type 3 vaccine is safe, infectious, immunogenic, and phenotypically stable in infants and children. J Infect Dis 199; 171: Karron RA, Makhene M, Gay K, Wilson MH, Clements ML, Murphy BR. Evaluation of a live attenuated bovine parainfluenza type 3 vaccine in two- to six-month-old infants. Pediatr Infect Dis J 199; 1: Clements ML, Belshe RB, King J, et al. Evaluation of bovine, cold-adapted human, and wild-type human parainfluenza type 3 viruses in adult volunteers and in chimpanzees. J Clin Microbiol 1991; 29: Belshe RB, Maassab HF, Mendelman PM. Influenza vaccine-live. In: Plotkin SA, Orenstein WA, eds. Vaccines. 4th ed. Philadelphia: WB Saunders, 2004: Channock RM. Parainfluenza viruses. In: Lennette EH, Schmidt NJ, eds. Viral, rickettsial and chlamydial infections. th ed. Washington, DC: American Public Health Association, 1979: Lee MS, Greenberg DP, Yeh SH, et al. Antibody responses to bovine parainfluenza virus type 3 (PIV3) vaccination and human PIV3 infection in young infants. J Infect Dis 2001; 184: Lee MS, Mendelman PM, Sangli C, Cho I, Mathie SL, August MJ. Half-life of human parainfluenza virus type 3 (hpiv3) maternal antibody and cumulative proportion of hpiv3 infection in young infants. J Infect Dis 2001; 183: Downloaded from at Pennsylvania State University on May 8, JID 200:191 (1 April) Greenberg et al.