444
Concise Communications
5. Ren R, Costantini F, Gorgacz E, Lee J, Racaniello V. Transgenic mice
expressing a human poliovirus receptor: a new model for poliomyelitis.
Cell 1990;63:353-62.
6. Koike S, Taya C, Kurata T, et al. Transgenic mice susceptible to poliovirus.
Proc Nat Acad Sci USA 1991;88:951-5.
7. Racaniello V, Ren R. Transgenic mice and the pathogenesis of poliomyelitis. Arch Virol 1994;9(suppl):79-86.
8. Dragunsky E, Gardner D, Taffs R, Levenbook I. Transgenic PVR Tg-I
mice for testing of poliovirus type 3 neurovirulence: comparison with
the monkey test. Biologicals 1993;21 :233-7.
9. Horie H, Koike S, Kurata T, et aI. Transgenic mice carrying the human
poliovirus receptor: new animal model for study of poliovirus neurovirulence. J Virol 1994;68:681-8.
10. Abe S, Ota Y, Koike S, et al. Neurovirulence test for oral live poliovaccines
using poliovirus-sensitive transgenic mice. Virology 1995;206:
1075-83.
JID 1997;175 (February)
11. Dragunsky E, Taffs R, Chernokhvostova Y, et al. A poliovirus-susceptible
transgenic mouse model as a possible replacement for the monkey
neurovirulence test of oral poliovirus vaccine. Biologicals 1996;24:7786.
12. van Steenis G, van Wezel AL, Sekhuis VM. Potency testing of killed
polio vaccine in rats. Dev Bioi Standard 1981;47:119-28.
13. Albrecht P, Enterline JC, Boone El, Klutch MJ. Poliovirus and polio
antibody assay in Hep-2 and Vero cell cultures. J Bioi Stand 1983; 11:
91-7.
14. Mahon BP, Katrak K, Nomoto A, Macadam AJ, Minor PD, Mills KH.
Poliovirus-specific CD4 + Th1 clones with both cytotoxic and helper
activity mediate protective humoral immunity against a lethal poliovirus
infection in transgenic mice expressing the human poliovirus receptor.
J Exp Med 1995; 181:1285-92.
15. Elisberg BL. Standardization of safety and potency tests of vaccines against
poliomyelitis Rev Infect Dis 1984;6(suppl):S519-22.
Dose Titration Study of Live Attenuated Varicella Vaccine in Healthy Children
Edward P. Rothstein, Henry H. Bernstein,
Pennridge Pediatric Associates,* Angela L. Ngai,
Iksung Cho, and C. Jo White
Temple University School of Medicine and St. Christopher's Hospital
for Children, Philadelphia; Merck Research Laboratories,
West Point, Pennsylvania
To approximate the effect of prolonged storage on safety and immunogenicity, healthy children
were given a single dose of the currently marketed live attenuated varicella vaccine (3625 pfu) or
of a partially heat-inactivated vaccine (1125 or 439 pfu), The 3 doses had similar antigen content
(attenuated plus inactive virus particles). The vaccine was well tolerated. No significant differences
in adverse reactions were observed. Although the seroconversion rates were excellent at each dose
(?:98%), the higher doses resulted in significantlygreater geometric mean antibody titers at 6 weeks
(10.5 and 10.6 ELISA U/mL) compared with the 439 pfu dose (5.7 ELISA U/mL), P ::0:; .01. One
year after immunization, differences in antibodies were similar to the 6-week postimmunization
results. Results indicate that until the date of expiry, the vaccine's immunogenicity will be preserved
and there will be no clinically important changes in type or frequency of adverse events.
The safety, immunogenicity, and efficacy of live attenuated
varicella vaccine, licensed in the United States for use in persons ~ 12 months old, has been demonstrated in> 10,000 study
participants. Routine administration of varicella vaccine to
healthy susceptible children has been recommended by the
American Academy of Pediatrics and the Advisory Committee
on Immunization Practices [1, 2].
Received 24 May 1996; revised 3 September 1996.
Informed consent was obtained from parents. Human experimentation guidelines of the US Department of Health and Human Services and those of the
Grand View Hospital Institutional Review Board were followed.
Financial support: Merck Research Laboratories.
Reprints or correspondence: Dr. Edward P. Rothstein, 711 Lawn Ave., Sellersville, PA 18960.
* Other members of Pennridge Pediatric Associates at the time of the study:
Ruth P. Schiller, Joseph A. C. Girone, Thomas J. Hipp, Ronald L. Souder,
Thomas I. Kennedy, and Deborah R. Faccenda.
The Journal of Infectious Diseases 1997; 175:444-7
© 1997 by The University of Chicago. All rights reserved.
0022-1899/97/7502-0027$01.00
In most previous studies, vaccine was prepared by filling a
single-dose vial with a specified number of plaque-forming
units (Pfu) or by diluting a known quantity of pfu to obtain
the desired immunizing dose [3-5]. Although these methods
vary the total immunizing dose, the ratio of attenuated to inactive viral antigen particles remains similar. These methods may
not accurately reflect the conditions that occur with prolonged
storage, when the antigen dose remains the same while the
ratio of attenuated to inactive virus particles decreases. This
study evaluated the immunogenicity and safety of the currently
marketed live attenuated varicella vaccine after partial heat
inactivation to approximate the effect of prolonged storage.
Methods
Subjects. Healthy 1- to 6-year-old children were enrolled in
the study from a private, mostly middle class, white, population.
Subjects had no history of varicella, zoster, or exposure to these
diseases for ?:4 weeks before immunization. Subjects were excluded if they were immunocompromised or had received immune
globulin or blood products within 3 months before immunization.
JID 1997; 175 (February)
Concise Communications
Vaccine. Live attenuated varicella vaccine (VARIVAX) was
derived from the Oka strain and developed by Merck Research
Laboratories as previously described [4, 5]. The vaccine used in
this trial (lot C-W591), produced using the same process as the
licensed vaccine, was one of five consistency lots manufactured
in 1991 [5]. To approximate the conditions of long-term storage
and reconstitution, 2 of 3 groups of single-dose vials containing
3625 pfu of lyophilized vaccine were exposed to a temperature of
32°C for 2 and 12 days, yielding vaccine that contained 1125
and 439 pfu/O.5 mL, respectively. Infectivity of the vaccine was
measured by plaque assay with a relative SD of 15%. The antigen
content of each preparation was similar before (0.9 U/O.5 mL) and
after heat inactivation (0.9-1.0 U/O.5 mL). Vaccine was stored at
- 20°C until reconstituted with sterile distilled water immediately
before administration.
Study design. After a medical history was obtained, children
were randomized to receive one 0.5-mL subcutaneous dose of
varicella vaccine containing 439, 1125, or 3625 pfu. Children were
not immunized within 1 month of another immunization or during
a febrile illness.
Signs and symptoms, including daily temperatures, were recorded by parents on a standardized reporting form for 42 days
after immunization. Parents were asked to notify the study physicians if their child developed a varicella-like rash at the injection
site or elsewhere, developed temperature ?:38.9°C, or had any
unusual reaction. Although the study was designed as "open,"
parents and study physicians were unaware of the dose of vaccine
given. Blood was obtained by venipuncture before immunization,
6 weeks later, and yearly thereafter. Routine serologic data are
available for 2 years.
Parents were requested to notify study physicians if they thought
their child developed chickenpox any time after immunization;
these children were seen by a study physician. If chickenpox was
even a remote possibility, a request for acute and convalescent
blood was made. If a child was suspected of having chickenpox
after immunization, yearly blood specimens were not obtained.
Breakthrough and exposure information was also collected when
participants returned for antibody persistence serum sampling.
Serologic assay. Serum was frozen until assayed at Merck
Research Laboratories. Humoral antibody to varicella-zoster virus
(VZV) was measured by a glycoprotein-based ELISA as previously described [6] and reported in ELISA units (EU). Pre- and
6-week postimmunization sera from each subject were assayed
simultaneously. One- and 2-year samples were tested separately
in the same laboratory using standardized assays. Suspected breakthrough chickenpox was considered confirmed if there was an
increase in antibody titer between acute and convalescent specimens of ?:4-fold, a convalescent antibody titer achieved a predetermined antibody value that correlated with breakthrough chickenpox, or both [7].. Laboratory personnel were unaware of the
subjects' vaccine assignments. Cellular immunity was not assessed.
Statistical analysis. All tests of significance were two-sided,
and results were considered significant at P ~ 05. No adjustment
was made for multiple pairwise comparisons. Geometric mean
(GM) antibody values were compared by analysis of variance, as
the log-transformed values appeared normally distributed. Proportions of subjects with 6-week antibody titers ?:5.0 EU/mL were
compared by Fisher's exact test. Proportions of subjects with ad-
445
verse events among groups were compared using X2 or Fisher's
exact test as appropriate. Due to the small sample size, nonsignificant P values do not necessarily imply similar immune responses.
Results
Demographics. Fifty children were enrolled at each dose
level; all completed the study to 6 weeks. At enrollment, the
mean age (2.9-3.2 years), gender, and percentage of children
initially seropositive (2%-8%) were similar for the 3 groups.
A total of 135 children completed the study for 1 year and 107
for 2 years; 88 and 85 subjects were actively surveyed for
breakthrough chickenpox in years 3 and 4, respectively.
Safety (table 1). The vaccine was generally well tolerated.
No serious adverse experiences occurred. No statistical differences in local or systemic adverse events, including oral temperatures ~38.9°C (10%-16%), were observed, and none were
considered clinically important.
Immunogenicity (table 2). Six weeks after immunization,
98%-100% of subjects in each group seroconverted. The GM
antibody responses of children who received 1125 or 3625 pfu
were significantly greater than in children who received 439
pfu (P ~ .01). The percentage of children achieving antibody
titers ~5.0 EU/mL was greater in the group given 3625 pfu
(P ~ .01) than in the group given 439 pfu. Although sera
for all children assayed 1 year after immunization remained
positive, the children who received the 2 higher doses had
higher GM antibodies than the children who received 439 pfu.
All sera assayed 2 years after immunization remained positive;
GM antibody values were 11.0, 21.6, and 12.1 EU/mL in the
groups receiving 439, 1125, and 3625 pfu, respectively. These
differences were not statistically different or clinically important.
Table 1. Percentage of healthy children with adverse events (without regard to causality) occurring within 42 days of immunization
with various doses of varicella vaccine.
Dose (Pfu)
439
1125
3625
(n == 50)
(n == 50)
(n == 50)
Local
Soreness
Ecchymosis
Erythema
Swelling
Varicella-like rash at injection site
10
8
2
2
0
0
18
12
0
6
2
0
12
8
0
2
0
4
Systemic
Temperature 37.7-38.8°C, orally
Temperature ~38.9°C, orally
Varicella-like rash
84
30
16
4
92
38
12
4
90
42
Adverse event
NOTE.
For all comparisons, P
> .05,
X2 or Fisher's exact test.
10
2
446
Concise Communications
JIO 1997; 175 (February)
Table 2. Antibody responses to various doses of varicella vaccine in initially seronegative children.
Dose (Pfu)
439
1125
3625
At 6 weeks (n)
Geometric mean (95% CI)
% seroconverting
% ;?:5.0 EU/mL (95% CI)
49
5.7 (4.1-8.1)
98
61 (46.2-75.1)
47
10.6* (7.9-14.2)
100
79 (63.9-89.8)
46
10.5* (8.5-13.0)
100
87* (73.0-95.9)
At 1 year (n)
Geometric mean (95% CI)
% seropositive
44
9.1 (5.6-14.8)
100
46
15.9 (12.0-21.0)
100
45
14.3 (10.9-18.8)
100
At 2 years (n)
Geometric mean (95% CI)
% seropositive
35
11.0 (6.6-18.5)
100
35
21.6 (12.4-37.5)
100
37
12.1 (8.4-17.6)
100
NOTE. CI, confidence interval; EU, ELISA units.
* Significantly >439 pfu, P ~ .01. Geometric mean compared by analysis of variance. Proportions of subjects
with 6 week antibody values ;?:5.0 EU/mL, Fisher's exact test.
Breakthrough varicella surveillance. During 4.3 years of
surveillance, parents of 15 children reported 16 episodes of
suspected breakthrough varicella. As the study physicians were
also the participants' primary care physicians, we believe most
breakthrough cases were captured. Six children with suspected
breakthrough varicella had received 439 pfu (6-week postimmunization titers: 1.6, 2.3, 5.1, 5.7, 6.8, and 13.0 EU/mL), 6
others received 1125 pfu (6-week postimmunization titers: 4.0,
4.4, 5.0, 11.9, 15.1, and 16.1 EU/mL), and 3 received 3625
pfu (6-week postimmunization titers: 1.5, 2.7, and 16.1 EUI
mL). Breakthrough cases occurred 14-40 months after immunization. All but 4 illnesses were seen by a study physician.
Each child with suspected chickenpox had mild disease (median number of maximum lesions and median number of maximum vesicles, 25 and 0, respectively), consistent with previous
reports [7-9]. Acute and convalescent serum specimens, available for 9 episodes of suspected chickenpox, confirmed the
disease in 7 children.
Another 3 children had only convalescent serum tested, but
comparison with previous surveillance titers was possible. In
the first child, 5 months after an antibody level of 1.5 EU/mL
at year 1, suspected chickenpox occurred. Six weeks later, his
antibody level was 39.5 EU/mL at year 2. In a second child,
7 weeks after a serum antibody level of 2.3 EU/mL at year 2,
suspected chickenpox occurred. The child's antibody titer was
67.5 EU/mL 2 months later. Although the >4-fold antibody
increase in these children suggests recent infection with VZV,
the lack of an acute serum specimen and the fact that both
convalescent values were lower than those usually associated
with recent breakthrough chickenpox in our patients (20011,563 EU/mL) and in other studies (558-5000 EU/mL) [7]
prevents definitive diagnosis. The third child developed suspected chickenpox 10 months after he had an antibody titer of
3.6 EU/mL; 3 weeks later, the level was 4213 EU/mL.
Discussion
In this study, heat inactivation rather than dilution was used
to generate decreased live attenuated virus titers while maintaining a constant antigen level. This method reflects more
accurately the amount of total antigen (attenuated virus plus
inactive virus particles) of the vaccine at expiry, when stored
and administered as directed. There were no statistically significant differences in the type or frequency of adverse events
with various live virus quantities over an 8-fold range or with
various ratios of attenuated to inactive virus. The frequency of
temperatures ~38.9°C was similar to that in a published double-blind placebo-controlled study in which ,...., 15% of children
who received the placebo had oral temperatures of ~38.9°C
during 58 days of follow-up [10].
Although a dose of 439 pfu resulted in excellent 6-week
seroconversions (98%) compared with doses of 1125 and 3625
pfu (each 100%), the 6-week OM antibody value was significantly lower in this group than in children who received higher
vaccine doses. The percentage of children achieving 6-week
antibody levels ~5.0 EU, a value roughly correlated with the
probability of developing chickenpox after immunization [8],
was also significantly lower in the group receiving 439 pfu
than in the group given 3625 pfu. Higher OM antibody values
were observed 1 year after immunization for the groups given
1125 and 3625 pfu of vaccine than in children given 439 pfu.
In year 2, subjects who received 1125 pfu had higher OM
antibody titers than the groups given 439 or 3625 pfu. These
differences, which probably reflect differences in number of
exposures, timing of exposure to natural disease, and blood
sampling, were not statistically different.
Breakthrough chickenpox occurred occasionally in recipients of all 3 doses. As this study was designed to assess only
safety and humoral immunogenicity, it was not possible to
JlD 1997; 175 (February)
Concise Communications
detect statistically significant differences in protection among
the 3 groups.
Our results were similar to those of a study of a prelicensure
varicella vaccine that was partially inactivated by exposure to
ambient temperature [11]. In that study, initially seronegative
children developed significantly lower humoral antibody levels
after receiving vaccine with similar antigen content but lower
attenuated virus content. Cell-mediated immunity (CMI) was
similar in the 3 groups 1 year after immunization. Although the
exact CMI mechanism involved in response to immunization,
infection, reinfection, reactivation of latent VZV, and booster
doses has not been defined, it is apparent that both humoral
and CMI responses play major roles in protecting against VZV.
In another study of seronegative children, 2 vaccine preparations that varied in antigen content but maintained similar pfu
were administered [12]. Six weeks after immunization, the humoral antibody response was similar, but the group receiving
vaccine with the higher antigen content had a greater CMI
response. The authors suggested that inactive virus particles
might prime for a response to replicating attenuated virus. In
contrast, in seropositive adults, the humoral and CMI responses
seem to be similar after attenuated vaccine (attenuated virus
plus inactive virus particles) or completely inactivated vaccine
containing a similar antigen content [13, 14]. From the results
of our study and of those cited, it seems that in seronegative
persons the immune response can be modulated by varying
the content of either attenuated virus or antigen, whereas in
seropositive subjects the immune response seems to be based
on antigenic content.
Results of this study provide reassurance that as long as the
vaccine is stored properly and administered within 30 min of
reconstitution, the minimum approved vaccine potency at the
expiration date (1350 pfu) will ensure that the vaccine's immunogenicity will be preserved and there will be no clinically
important changes in type or frequency of adverse events.
While our study documents that doses of 439 pfu are immunogenic, doses containing 1125 or 3625 pfu resulted in higher GM
antibody values. Our results suggest that to ensure adequate
immunogenicity the vaccine should contain ~ 1125 pfu of live
attenuated varicella virus at the time of administration.
Acknowledgments
We thank Christina Chan, Joseph Eiden, Barbara Watson, and
Ann Arvin for thoughtful comments; Lynne DeSilvis and other
447
nursing staff of Pennridge Pediatric Associates; the parents and
children who participated in this study; Sharon Rogin for data
handling support; and William Humi and Beverly Rich, Merck
Research Laboratories, for doing ELISAs.
References
1. Committee on Infectious Diseases, American Academy of Pediatrics. Recommendations for the use oflive attenuated varicella vaccine. Pediatrics
1995; 95:791-6.
2. Centers for Disease Control and Prevention. Varicella vaccine. MMWR
Morb Mortal Wkly Rep 1995;44:264.
3. Johnson CE, Shurin PA, Fattlar D, Rome LP, Kumar ML. Live attenuated
varicella vaccine in healthy 12- to 24-month old children. Pediatrics
1988;81:512-8.
4. White CJ, Kuter BJ, Hildebrand CS, et al. Varicella vaccine (VARlVAX)
in healthy children and adolescents: results from clinical trials, 1987 to
1989. Pediatrics 1991; 87:604-10.
5. Ngai AL, Staehle BO, Kuter BJ, et al. Safety and immunogenicity of one
vs. two injections of Oka/Merck varicella vaccine in healthy children.
Pediatr Infect Dis J 1996; 15:49-54.
6. Wasmuth EH, Miller WJ. A sensitive enzyme-linked immunosorbent assay
(ELISA) for antibody to varicella-zoster virus (VZV) using purified
VZV glycoprotein antigen. J Med ViroI1990;32:189-93.
7. Watson BM, Piercy SA, Plotkin SA, Starr SE. Modified chickenpox in
children immunized with the Oka/Merck varicella vaccine. Pediatrics
1993; 91: 17-22.
8. White CJ, Kuter BJ, Ngai A, et al. Modified cases of chickenpox after
varicella vaccination: correlation of protection with antibody response.
Pediatr Infect Dis J 1992; 11:19-23.
9. Bernstein HH, Rothstein EP, Pennridge Pediatric Associates, et al. Clinical
survey of natural varicella compared with modified varicella after immunization with live attenuated Oka/Merck varicella vaccine. Pediatrics
1993; 92:833-7.
10. Weibel RE, Neff BJ, Kuter BJ, et al. Live attenuated varicella virus vaccine. Efficacy trial in healthy children. N Engl J Med 1984;310:140915.
II. Watson B, Piercy S, Soppas D, et al. The effect of decreasing amounts of
live virus, while antigen content remains constant, on immunogenicity
of Oka/Merck varicella vaccine. J Infect Dis 1993; 168:1356-60.
12. Bergen RE, Diaz PS, Arvin AM. The immunogenicity of the Oka/Merck
varicella vaccine in relation to infectious varicella-zoster virus and relative viral antigen content. J Infect Dis 1990; 162:1049-54.
13. Sperber SJ, Smith BV, Hayden FG. Serologic response and reactogenicity
to booster immunization of healthy seropositive adults with live or
attenuated varicella vaccine. Antiviral Res 1992; 17:213- 22.
14. Hayward AR, Buda K, Levin MJ. Immune response to secondary immunization with live or inactivated VZV vaccine in elderly adults. Viral
Immunol 1994; 7:31-6.