Early Loss of Passive Measles Antibody in Infants of Mothers With Vaccine-Induced Immunity

(1)

Early

Loss

of

Passive

Measles

Antibody

in

Infants

of

Mothers

With

Vaccine-Induced

Immunity

Yvonne A. Maldonado, MD*; Elizabeth C. Lawrence, BA*; Ross DeHovitz, MDX;

Harry Hartzell, MDX; and Paul Albrecht, MD

ABSTRACT. Background. Maternally derived passive

measles antibody may interfere with vaccine-induced

immunity in infants less than 12 months of age.

How-ever, early loss of passive measles antibody may occur in infants of women who received measles vaccine because measles vaccine induces lower antibody titers than does natural infection.

Methods. Persistence of passive neutralizing measles antibody was studied longitudinally in a group of nor-mal infants as a function of maternal measles titer at birth and maternal date of birth. Maternal serum and

cord blood specimens were tested from 162 women and

their newborns, from 51 of these infants at 9 months of age and from 63 at 12 months of age.

Results. Seventy-one percent of sera from 9-month-old infants (36 of 51, 95% confidence interval 68% to 84%) and 95% of samples from 12-month-old infants (60 of 63, 95% confidence interval 89% to 101%) had no detectable neutralizing measles antibody. Measles geometric mean

titers were significantly higher at delivery in mothers

whose infants were seropositive at 9 and 12 months

com-pared with mothers whose infants were seronegative at 9

and 12 months. All infants with detectable measles anti-body at 9 or 12 months had mothers born before 1963, before the vaccine era, and both maternal and cord blood measles geometric mean titers decreased significantly with decreasing maternal age.

Conclusions. Persistence of passive measles antibody is uncommon by 12 months of age; earlier antibody loss is related to lower maternal age and maternal measles

titer. Pediatrics 1995;96:447-450; measles, vaccine, passive

immunity, antibody, immunizations.

ABBREVIATIONS. ACIP, Advisory Committee on Immunization

Practices; HI, hemagglutination inhibition; GMT, geometric mean

titer.

From the *Depag.fient of Pediatrics, Stanford University School of

Medi-cine, Stanfonl, California; jPalo Alto Medical Foundation Clinic, Palo Alto, California; and the §Division of Viral Products, CBER, Food and Drug Administration, Bethesda, Maryland.

This study was conducted with approval from the Committees for the

Protection of Human Subjects at Stanford University and the Palo Alto Medical Foundation. Informed consent was obtained from all study

partic-ipants or their parents. We are indebted to the pediatricians at the Palo Alto

Medical Foundation Clinic for assistance in recruitment of study patients, and to Barbara Sullivan and Marcia Chmyz for technical assistance with this study.

Received for publication Jun 6, 1994; accepted Nov 21, 1994.

Reprint requests to (YAM.) Room G312, Department of Pediatrics, Stan-ford University School of Medicine, Stanford, CA 94305.

Interference by passive measles antibody has been considered the primary obstacle to the successful immunization of young infants with live, attenuated

measles vaccine. Until recently, the United States

Public Health Service Advisory Committee on

Im-munization Practices (ACIP) recommended routine

measles vaccination at the age of 15 months, with a

second dose of measles vaccine before school entry.’

This strategy was based on studies conducted in the

1970s, which documented a reduction in the

immu-nogenicity of measles vaccine and diminished

clini-cal protection against measles in infants vaccinated before 12 months of age.29 Failure rates of up to 95% were reported in infants vaccinated before 12 months

of age, and were inversely proportional to age at

vaccination. Yeager et aP#{176}noted that vaccine failure in infants correlated not only with age at vaccination, but also with higher maternal measles titers, as mea-sured by hemagglutination inhibition (HI).

Neutral-izing measles antibody assays using low-passage

virus pools were later demonstrated by Albrecht

et al”12 to be approximately 60 times more sensitive

than HI assays in detecting preexisting measles

an-tibody in infants less than 12 months of age who

were measles vaccine failures. The titer of passive (maternal) neutralizing measles antibodies also cor-related with seroconversion rates and

postimmuni-zation geometric mean titers (GMTs) in infants

receiving measles vaccine at 12 months of age.’3 Although these observations provided a coherent rationale for giving measles vaccine at 15 months, the effect of passive, transplacentally derived maternal

antibody on measles vaccine immunogenicity has

been reexamined because passive maternal antibody

among infants born in the 1970s was derived from

women whose measles immunity was due to natural

infection. Measles vaccine was licensed for use in the U.S. in 1963, and the majority of women of reproduc-tive age in this country now have measles

vaccine-induced immunity. Because measles vaccine induces

lower antibody titers than does natural infection, lower levels of passively acquired maternal measles

antibody might be expected among infants whose

maternal immunity is vaccine derived.’4-’5 Published

studies conducted outside the U.S. support this

hy-pothesis.1618 Based on these observations, recently

modified recommendations from the ACIP include

the administration of measles vaccine as early as 12

months of age and any time between 12 and 15

months of age.19 These recommendations were based

at Viet Nam:AAP Sponsored on September 1, 2020 www.aappublications.org/news

(2)

RESULTS

A cohort of 162 mother-infant pairs was enrolled in the study. Seventy-three of the mothers were born

before 1958, 60 were born between 1958 and 1962,

and 29 were born after 1962. Maternal blood was

available from 145 women, cord blood from 141 of

their infants, and blood samples from 80 of the

in-fants. Fifty-one blood samples were available from

the infants at 9 months of age and 63 samples from

infants at 12 months of age. No serum samples were

available from 82 of the infants. To determine

whether a selection bias existed in the group of

in-fants for which 9- or 12-month serum samples were

available, we compared infants who had serum

sam-ples available at 9 or 12 months and those who had

no serum samples available at 9 or 12 months. No

significant differences were found in maternal birth

year or maternal measles GMT at delivery between

the groups.

The GMTs from maternal and cord sera were

grouped by maternal birth year (Fig 1). Maternal age

groups were compared based on the likelihood of

measles infection or measles vaccination in

child-hood; the maternal birth year comparisons were:

<1958 vs 1958, <1958 vs >1962, and 1962 vs

>1962. Although measles GMTs from cord blood

were consistently higher than maternal GMTs, both

maternal and cord blood measles GMTs were

signif-icantly lower as maternal age decreased. To

deter-mine whether significant differences existed in

ma-ternal measles titer based on the presumed source of

maternal immunity, we compared maternal GMTs

among three categories based on maternal birth year.

Differences in GMT between all categories showed

statistical significance.

Measles neutralizing antibody titers in 9- and

12-month-old infants were also grouped by maternal

birth date. All infants who were still seropositive at 9

months (n = 15) or 12 months (n = 3) had mothers

born before 1963 (Fig 2); the persistence of measles

antibody at 9 months of age was significantly more

likely among infants with maternal birth dates before

5000 4500 4000 I-I- 3500 z 4 w C.) I-w 0 ‘U C, 000B<1958 #{149}DOB>1962 3000 --___________ 2500 2000 1500 1000 500 0 MATERNAL* CORD

448 EARLY LOSS OF PASSIVE MEASLES ANTIBODY

on unpublished data presented to the ACIP, and no

published data have been available to compare the

duration of passive measles immunity from cohorts

of U.S.-born women with vaccine- versus

wild-type-induced immunity. We sought to evaluate

prospec-tively the loss of passive neutralizing measles

anti-body among infants in the U.S. as a function of

whether maternal immunity was vaccine-induced or

due to natural infection.

Study Population

METHODS

We selected a suburban population with no recent measles

transmission to identify differences in the persistence of passive measles antibody in infants of U.S-born women with wild-type versus measles vaccine-induced immunity. The Palo Alto Medical Foundation Clinic borders two counties, San Mateo and Santa Clara, each with populations of greater than 500 000. Since 1972,

Santa Clara county has reported fewer than 150 measles cases annually except for 1975, when 1031 measles cases were reported in the county. San Mateo county has reported fewer than 40 cases annually since 1974, except for 1975, when 493 cases were

re-ported. Maternal birth date was used as a proxy for source of maternal immunity. Mothers born before 1958 were presumed to have acquired measles infection, as most persons in the U.S. aged

10 years or older in 1967 had acquired natural measles infection. Mothers born after 1962 were presumed for the most part to have acquired immunity by measles immunization, as measles vaccine

was licensed for use in 1963, and disease incidence in the U.S. dropped almost 90% by 1966 to 1968 as a result of widespread vaccine use. Mothers born between 1958 and 1962 were likely to be

a mixed group, some having experienced measles infection and some having acquired immunity through immunization.

Human subjects approval was obtained for this study from the

Stanford University Committee for the Protection of Human

Sub-jects and the Palo Alto Medical Foundation Clinic. Women

deliv-ering at Stanford University Hospital between May 1, 1991 and

September 30, 1991 and followed at the Palo Alto Medical Clinic

were eligible to participate in the study if they were born in the

U.S. and if their infants were born after 36 weeks’ gestation

weighing more than 2000 g. If consent was obtained, maternal and cord blood samples were obtained at the time of delivery. Cord blood samples were obtained by clamping and cutting the

umbil-ical cord and withdrawing blood from one of the three umbilical

vessels. The cord was not milked or stripped. Blood was then

collected from infants during 9- and 12-month well child visits at

the Palo Alto Medical Clinic. Sera were stored at -20#{176}C.

Laboratory Procedures

The test for measles neutralizing antibody was performed by a modification of the previously described assay.” Briefly, the con-ventional plaque reduction neutralization was carried out in

plas-tic 24-well plates using Vero cell monolayers. Serial dilutions of

heat-inactivated serum (56#{176}Cfor 30 minutes) were mixed with an equal volume containing 25 to 35 plaque-forming units of a low-passage strain of Edmonston measles virus and incubated for I

hour and 45 minutes at 36#{176}Cin a 5% CO2 incubator. Ihe inoculum

was then replaced with 1.0 mL of overlay containing 2.0% car-boxymethylcellulose in Leibovitz medium, with 5% fetal bovine

serum and antibiotics, and incubated for 4 days at 36#{176}C.On day 4,

0.5 mL of 0.04% neutral red in Eagle’s minimal essential medium

was added to each well, and the test was incubated for one more

day. On day 5, the overlay was removed and the wells were fixed

with 10% formaldehyde to enhance and preserve the staining.

Antibody titers were calculated as the dilution of serum reducing

plaques by 50%. A titer of less than 1:8 was considered negative and assigned a value of I :2 for statistical purposes.

Data Analysis

The reciprocals of measles neutralization titers were

trans-formed logarithmically, and GMIs were calculated. Differences

between groups were determined by Fisher’s exact test or x. and

differences in antibody titers were determined by the

Mann-Whit-ney test (Kruskal-Wallis), with statistical significance at P .05.

Fig 1. Geometric mean titers of measles neutralizing antibodies in

maternal and cord blood, grouped by maternal birth date. Values

of less than I :8 are considered negative and assigned a value of 1:2. DOB, maternal date of birth. values for differences in

maternal GMT based on maternal birth dates, by Fisher’s exact

test: P = .000002 for the difference between maternal birth date

<1958 and >1962; P < .0000001 for the difference between maternal

birth date <1958 and 1958; P = .0002 for the difference between

maternal birth date s1962 and >1962.

(3)

1000 I->. 0 0 100 z 4 C, z N 1 10 I-‘U z (I) IU -J C,) 4 ‘U .

i

!

S ,

m:

<1958* 1958-1962 >1962 <1958 1958-1962 >1962

9 MONTh OLD INFANTS 12 MONTh OLD INFANTS

MATERNAL BIRThDATE

Fig 2. Measles neutralizing antibody titers in infants at 9 and 12

months of age, grouped by maternal birth date. Values of less than

1 :8 are considered negative and assigned a value of I :2. *1) .05

for difference between maternal birth date <1958 and >1962,

Mann-Whitney LI test for two groups.

1958 than in those with maternal birth dates after

1962 (P = .05, Fisher’s exact test). A statistical

com-parison at 12 months of age could not be made

because the overall seropositive rate at 12 months was low. Overall, 36 of 51 samples (71

%,

95% confi-dence interval 68% to 84%) from 9-month-old infants

had no detectable measles antibody, and 60 of 63

samples (95%, 95% confidence interval 89% to 101%)

from 12month-old infants had no detectable measles

antibody. Among 9-month-old infants with

detect-able measles antibody (n = 15), the GMT was 48

(range 9 to 241), and among 12-month-old infants (n

= 3), the GMT was 37 (range 24 to 86). The measles

GMT for mothers of infants who were seronegative

at 9 months (693, 95% confidence interval 436 to

1104) was significantly lower than the measles GMT

for mother of infants who remained seropositive at

9 months (3048, 95% confidence interval 1527 to

6081).

DISCUSSION

This study is one of the first to document a signif-icant association between early loss of passive mea-sles antibody in the first year of life and maternal

birth year among infants of women born in the

United States. Persistent seropositivity at 9 and 12

months of age occurred only among infants whose

mothers were born before 1963. In addition, maternal

and cord measles titers decreased significantly by

decreasing maternal age group. These results

mdi-cate that the duration of passive measles immunity is related to the maternal source of immunity. Mothers born before 1958 were likely to have been exposed to

natural measles infection, with resulting high

neu-tralizing measles antibody titers. Thereafter,

expo-sure to natural measles infection in the U.S.

de-creased steadily, and the use of measles vaccination

became widespread beginning in 1963. Our study

demonstrated that infants of women born before

1958 are more likely to have persistent passive

mea-sles antibody at 9 and 12 months of age, whereas

infants of women born after 1962 were seronegative

by 9 months. Infants of mothers born between 1958

and 1962 are at intermediate risk of having persistent

passive measles antibody at 9 and 12 months of age.

In the earliest detailed investigations of the

hu-moral immune response to measles vaccine in young

infants, Krugman et al.2#{176}used the HI method to

detect measles antibody and concluded that clinical protection did not necessarily correlate with HI

ti-ters. Techniques for quantitating neutralizing

mea-sles antibody have been improved significantly with

the development of a highly sensitive plaque reduc-tion neutralization method,” which is at least 50- to

60-fold more sensitive than conventional HI

methods.

Studies conducted outside the U.S. support our

finding of associations between young maternal age,

lower maternal measles titers, and the predominance of measles-seronegative infants at 9 and 12 months of age. For instance, Pabst et al16 found significantly

lower measles neutralizing antibody titers among

Canadian women born after 1964 with documented

measles vaccination compared with women born

be-fore 1958 with presumed measles-induced

immu-nity. By 7 months of age, 90% of infants of the

younger mothers and 65% of infants of the older

mothers had no detectable measles neutralizing

an-tibody. Jenks et al’7 conducted a cross-sectional study

in England and found lower measles antibody titers

among infants whose mothers had received measles

vaccine compared with unvaccinated mothers, with

immunity presumed to be due to natural infection.

Lennon et al’8 evaluated neutralizing measles GMTs

among American women by year of birth. Measles

GMT declined by birth year, with a sharp decline

among women born between 1955 and 1961. Our

study supports the predictions of Lennon et al that

95% of infants of mothers born after 1963 would be

seronegative by 8.5 months of age and that 2% or less

of all these infants at 12 months of age would have

persistent maternal antibody.’8 Finally, in a yet

un-published prospective study in Los Angeles,

Markowitz et al2’ found that mothers born before

1957 had higher measles neutralizing antibody titers

than mothers born in 1957 or later; infants of the

younger mothers were more likely to lack measles

antibody at 6, 9, and 12 months compared with

in-fants born to the older mothers.

Women in this study came from a predominantly

middle-class, suburban population, which might not

be representative of other groups for whom measles

immunization is recommended. For instance,

women from urban inner-city populations might

have had recent exposure to measles infection during

the epidemics of the mid-1980s and would have

de-veloped high measles titers regardless of immuniza-tion status, through either primary or anamnestic antibody responses to measles infection. Also,

immi-grant populations might have been exposed to

mea-sles infection before arrival in the U.S. and may have

subsequently developed high measles antibody

ti-ters. In these populations, maternal measles antibody titers and duration of passive immunity in their

in-fants might be elevated compared with those

mea-sured in our study population. However, the extent

of boosting experienced during the epidemic period

between 1986 and 1991 would be limited primarily to

selected inner-city populations, where these

(4)

450 EARLY LOSS OF PASSIVE MEASLES ANTIBODY ics were concentrated. In addition, there is evidence

that the duration of passive measles antibody in

infants of women from developing areas of the world is lower than among infants from developed areas,15

and therefore infants born to immigrant women

might not be expected to have a longer duration of

passive measles immunity compared with infants in

our study population.

Our study demonstrates that widespread loss of

maternal measles antibody may occur by 9 months of

age among infants of U.S.-born women, especially

among infants of mothers with presumed

vaccine-induced measles immunity. Our results indicate that

by 12 months of age, persistence of passive measles antibody is rare regardless of the maternal source of

immunity. The optimal schedule for measles

vacci-nation of young infants in the U.S. would balance the risk of early loss of maternal antibody in the majority of U.S. infants with the risk of primary vaccine

fail-ure due to passive measles immunity. In practical

terms, because approximately 83% of the 1994

an-nual birth cohort may be expected to have mothers

born after 1963, instituting measles vaccination at

12 months of age for all infants of women born after 1963 would exclude 17% of the birth cohort. In fact, if all infants in the 1994 birth cohort were vaccinated

at 12 months of age, then approximately 13 000

in-fants, or only 0.4% of the annual birth cohort, would be predicted to have low-level persistent passive an-tibody at the time of vaccination, based on our anal-ysis of antibody persistence in infants whose

moth-ers were born before 1963. Our data indicate that

more than 99% of the annual birth cohort will be

susceptible to measles infection at 12 months of age. This study supports the recent ACIP recommenda-tion to extend the age of measles vaccination to 12 to

15 months of age. However, if no change from the

current policy recommending measles vaccination

through 15 months of age is instituted, then some

infants will still be susceptible to measles infection

for at least 3 to 6 months before immunization. This interval of susceptibility is of major concern because the morbidity and mortality associated with measles infection is most severe among infants. The results of

this study provide compelling evidence to support

the uniform institution of measles vaccination by 12

months of age, at least for populations in which

measles vaccine is the primary source of maternal

measles immunity. Our study indicates an even

longer period of susceptibility, from at least 9 months to 15 months of age, for up to 70% of all infants and

for an even larger proportion if one considers only

those infants whose mothers have vaccine-induced

immunity. Because the extent of primary measles

vaccine failure is unknown in infants less than 12

months of age in the absence of passive maternal

antibody, it is possible that the ability of young

in-fants to respond to measles vaccine may be limited

by mechanisms other than those related to passive

neutralizing measles antibody. The importance of

evaluating the immunogenicity of measles vaccine in

seronegative infants between 9 and 12 months of age

should therefore be a priority in optimizing measles vaccination strategies.

REFERENCES

I. Centers for Disease Control. Measles prevention: recommendations of the Immunization Practices Advisory Committee (ACIP). MMWR. 1989; 38:1-18

2. Liimemann CC, Rotte TC, Schiff GM, Youtsey JL. A seroepidemiologic study of a measles epidemic in a highly immunized population. A,;i I Epideniol. 1972;95:238-246

3. Wilkins j, Wehrle PF, Portnoy B. Live, further attenuated rubeola vac-cine. A,n IDis Child. 1972;123:190-192

4. Schluederberg A, Lamm SH, Landrigan PJ, Black FL. Measles immunity in children vaccinated before one year of age. An, IEpidemiol. 1973;97: 402-409

5. Krugman RD. Rosenberg R, McIntosh K, et al. Further attenuated live measles vaccines: the need for revised recommendations. I Pediatr.

1977;91 :766-767

6. Krugman S. Present status of measles and rubella immunization in the United States: a medical progress report. JPediatr. 1977;90:1-12

7. Shasby DM, Shopfe TC, Downs H, Herrmann KL, Polkowski J.

Epi-demic measles in a highly vaccinated population. N EngI JMed. 1977; 296:585-589

8. Shelton JD, Jacobson JE, Orenstein WA, Schulz KF, Donnell HD. Mea-sles vaccine efficacy: influence of age at vaccination vs. duration of time since vaccination. Pediatrics. 1978;62:961-964

9. Wilkins J, Wehrle PF. Additional evidence against measles vaccine

administration to infants less than 12 months of age: altered immune response following active/passive immunization. I Pediatr. 1979;94: 865-869

10. Yeager AS. Davis JH, Ross LA. Harvey B. Measles immunization-successes and failures. JAMA. 1977;237:347-351

I I. Albrecht P, Herrmann K, Burns GR. Role of virus strain in conventional

and enhanced measles plaque neutralization test. IVirol Methods. 1981; 3:251-260

12. Orenstein WA, Albrecht P. Herrmann KL, Bernier R, Bart KJ, Rovira EZ.

The plaque-neutralization test as a measure of prior exposure to

mea-sles virus. IInfect Dis. 1987;155:146-149

13. Albrecht P, Ennis FA, Saltzman EJ, Krugman S. Persistence of maternal antibody in infants beyond 12 months: mechanism of measles vaccine failure. JPediatr. 1977;91:715-718

14. Markowitz LE, Preblud SR. Fine PEM, Orenstein WA. Duration of live measles vaccine-induced immunity. Pediatr Infect Dis I.1990;9:101-110 15. Black FL. Measles active and passive immunity in a worldwide

per-spective. Prog Med Virol. 1989;36:1-33

16. Pabst HF, Spady DW, Marusyk RG, et al. Reduced measles immunity in infants in a well-vaccinated population. Pediatr Infect Dis I. 1992;11: 525-529

17. Jenks P1, Caul EO, Roome APCH. Maternally derived measles immu-nity in children of naturally infected and vaccinated mothers. Epidemiol Infect. 1988;101 :473-476

18. Lennon JL, Black FL. Maternally derived measles immunity in era of vaccine-protected mothers. JPediatr. 1986;108:671-676

19. Centers for Disease Control. General recommendations on immuniza-tions. Recommendations of the Immunization Practices Advisory Corn-mittee (ACIP). MMWR. 1994;43:1-38

20. Krugman 5, Giles JP, Friedman H, Stone S. Studies on immunity to measles. IPediatr. 1965;66:471-488

21. Markowitz LE, Albrecht P. Demonteverde R, Rhodes P, Patriarca P.

Kaiser Measles Vaccine Study Group. Declining measles antibody titers in mothers and infants, United States. Abstract 423, 32nd Interscience Conference on Antimicrobial Agents and Chemotherapy. Anaheim, Califor-nia, October 11-14, 1993

22. National Center for Health Statistics. Supplements to the Monthly Vital Statistics Report: Advance Reports 1987. Vital and health statistics, series 24, no. 4. Washington, DC: Public Health Service; 1990

(5)

1995;96;447

Pediatrics

Albrecht

Yvonne A. Maldonado, Elizabeth C. Lawrence, Ross DeHovitz, Harry Hartzell and Paul

Immunity