RESPIRATORY SYNCYTIAL VIRUS DISEASE IN INFANTS …

AHtRidif JoTmxii. or EpronnoiiOOT Vol. 89, No. 4Copyright O 1883 by The Johns Hopkins University PrintM in US.A.

RESPIRATORY SYNCYTIAL VIRUS DISEASE IN INFANTSDESPITE PRIOR ADMINISTRATION OF ANTIGENIC

INACTIVATED VACCINE1- 2

HYUN WHA KIM, JOSE G. CANCHOLA', CARL D. BRANDT, GLORIA PYLES,ROBERT M. CHANOCK, KEITH JENSEN, AND ROBERT H. PARROTT4

(Received for publication August 8, 1968)

Kim, H. W. , J. G . Canchola, C. D. Brandt, G . Pyles, R. M. Chanock, K. Jensen

and R. H. Parrott (Children's Hosp. of D.C, Wash., D.C. 20009) . Respiratory

syncytial virus disease in infants despite prior administration of antigenic in-

activated vacdne. Amor. J. Epid., 1969, 89: 422-434 .—In response to three

injections of alum precipitated, 100X concentrated, formalin inactivated RS

vaccine (tot 100), 4 3 % of infant vaccinees displayed a 4-fold or greater rise in

serum neutralizing antibody and 9 1 % displayed a 4-fold or greater rise in

serum CF antibody. When RS virus became prevalent in the community, the

rate of RS virus infection in infants who received this vaccine was not remarkably

different from that in control infants who received parainfluenza vaccines. How-

ever, 8 0 % of RS vaccinees required hospitalization at the time of RS infection

whereas only 5 % of such infections among parainfluenza vaccinees resulted in

admission to the hospital. Illnesses among the RS vaccinees who underwent

natural infection included pneumonia, bronchiolitis, and bronchiolitis with pneu-

monia in a majority and rhinitis, pharyngitis and bronchitis in a few. It seems clear

that infants who received this vaccine were not protected against natural in-

fection and also, when they became naturally infected their illness was more

severe than that seen in cohorts who received a similar parainfluenza type 1

vaccine. These findings indicate that vaccine-induced RS virus serum antibody

alone does not protect against illness and suggest that serum antibody without

local respiratory antibody may play a part in the production of disease. W e have

also observed that the highest incidence of serious RS virus illness occurring

naturally is under six months of age when maternally derived serum antibody is

present. These findings together suggest that RS virus illness in infants is an

immunologic phenomenon wherein the virus and serum antibody interact to

produce severe illness.

antibodies; immunity; immunology; pneumonia; respiratory syncytial virus;

respiratory tract diseases; vaccines; viral vaccines

There seems little doubt that the respira- most important respiratory tract pathogentory syncytial (RS) virus is the single for infanta and children (1-12). Infection

with this agent is prevalent for a three to1 These studies supported by National Institute

of Allergy and Infectious Diseases Contract No. 'Present address: Department of Microbiology,PH^3-62-473 and Grant No. AI01528-13. University of Texas at San Antonio, San Antonio,

'From the Research Foundation of the Chil- Texas.dren's Hospital of the District of Columbia and 'Reprint requests to Dr. Parrott, Children'sGeorgetown University School of Medicine, Wash- Hospital of the District of Columbia, 2126 Thir-ington, D.C; Laboratory of Infectious Diseases, teenth St, N.W., Wash., D.C, 20009; or to theNational Institutes of Health, Bethesda, Mary- Vaccine Development Branch, National Instituteland; and Chas. Pfizer and Co, Groton, Connecti- of Allergy and Infectious Diseases, N I B , Be-cut thesda, Md. 20014.

422

HSV ILLNESS IN INFANTS DESPITE INACTIVATED VACCINE 423

four month period every year in virtuallyall pediatric populations in which it hasbeen studied, and when it is prevalent inthe community, hospitalization for bron-chiolitis and bronchopneumonia among in-fants reaches a peak (4). An effectivevaccine against early infection with thisagent could be expected to produce a sig-nificant reduction in lower respiratory tractmorbidity and hospitalization among in-fants and children. Initial efforts at im-munoprophylaxis have been concentratedprimarily on the development of an in-activated vaccine. Progress in the develop-ment of a live attenuated vaccine has pro-ceeded at a slower pace since susceptiblelaboratory animals do not develop pul-monary pathology nor are genetic markersavailable which correlate with virulence.

A successful inactivated RS virus vac-cine must satisfy certain unusual require-ments. Epidemiologic and serologic evi-dence indicates that moderate levels ofserum neutralizing antibody do not pro-vide effective protection against seriousRS virus disease during early infancy (13).In addition, the most serious RS virus ill-nesses occur in early infancy and there-fore protection must be provided in thefirst few months of life at a time when in-fants possess maternally transmitted serumantibody (13). Thus the goal of an im-munoprophylactic program for infantsmust be either to stimulate serum antibodylevels higher than those provided bymaternal transfer or alternately to inducethe development of local respiratory tractantibody. Attainment of the former goalwith inactivated vaccine, however, is madedifficult by the poor yield of RS virus intissue culture and the poor antigenicity ofRS viral antigens.

As a part of a collaborative effort in vac-cine development we have tested the anti-genicity in infants of a variety of in-activated experimental vaccines for RSand parainfluenza viruses. We have limitedour studies to infants less than one year ofage, since primary infection with RS and

type 3 parainfluenza viruses and associatedsevere illness occur most often during thisperiod. This report will describe thefailure of an antigenic inactivated respira-tory syncytial virus vaccine to provide pro-tection against infection and illness. Inaddition, we will describe an altered re-sponse of infant vaccinees in whom illnessassociated with naturally acquired RS in-fection was of greater than usual severity.

MATERIALS AND METHODS

Vaccines

In the present study, three different vac-cines were evaluated. 1) A respiratorysyncytial virus vaccine (lot 100) wasprepared from the Bernett strain whichwas grown in vervet monkey kidney tis-sue culture. This virus was isolated inprimary human embryonic kidney cul-tures from a throat washing at the Na-tional Institutes of Health in May, 1961,and was passaged three additional times inprimary human embryonic kidney culturesand 10 times in vervet monkey kidney cul-tures. A 1:50 dilution of this seed viruswas inoculated into vervet monkey kidneybottle cultures which were maintained withEagle's basal medium in Earle's salt solu-tion without serum. Early harvests werecollected after four days incubation at36 C when 10 to 25 per cent of the cellsexhibited a cytopathic effect; late harvestswere collected after the cultures were heldat 36 C for seven days and had developed50 to 90 per cent cytopathic effect.

The various sublot tissue culture fluidharvests were clarified individually byfiltration through Millipore SM mem-branes and inactivated with formalin(1:4,000 final) at 36 C for 72 hours. Thefluids then were centrifuged at 50,000 rpmin a Sharpies rotor and the resultant pelletwas resuspended in Eagle's basal mediumin Earle's salt solution to give a con-centration factor of 25-fold. A further 4-fold concentration was produced by pre-cipitation with alum (4 mg/ml) and

424 KTM ET All.

resuspension in one-fourth the originalvolume of Eagle's medium, thus achievinga final concentration factor of 100-fold.Polymyxin B, neomycin, and streptomycinat 200 units/ml each and benzethoniumchloride preservative (1:40,000 final) wereadded. At the conclusion of potency test-ing and completion of teste for possibleadventitious agents, three sublots fromearly harvests and 11 sublots from lateharvests were pooled to prepare vaccinelot 100. Potency tests were carried out inguinea pigs and cynomologous monkeys,which were tested for complement-fixing(CF) and neutralizing antibody after vac-cination. Antigen extinction endpointswere determined in guinea pigs which wereinjected intraperitoneally with 0.5 ml of avaccine dilution, given another similar in-jection seven and 14 days later and bledseven days after the last injection. The50 per cent antigenic extinction CF anti-body endpoint in guinea pigs was greaterthan 1:40 on one occasion and 1:64 onanother.

The safety test protocol was based uponthe requirements of the Division of Bio-logics Standards for an inactivated mon-key kidney tissue culture derived measlesvaccine (14). Safety tests were conductedin cynomologous monkeys, guinea pigs,rabbits and mice, with satisfactory results.Tests in tissue cultures for adventitiousagents, including SV40, were negative.Vaccine lot 100 was packaged in 3.0 mlvaccine vials with 3.4 ml fills.

2) A parainfluenza type 1 virus vaccine(lot 23) was prepared from an isolate ob-tained from an infant with croup by es-sentially the same procedure as the RSvirus lot 100 vaccine. The virus was iso-lated and propagated in human embryonickidney culture. Subsequently, it was grownin vervet monkey kidney tissue cultureand this system was used for the prepara-tion of vaccine. After millipore nitrationand formalin inactivation, tissue culturefluid harvests were concentrated 25-foldby centrifugation and then 4-fold further

by alum precipitation. The final concentra-tion was 100-fold.

3) A trivalent aqueous parainfluenza vi-rus vaccine (lot 6279) was prepared fromamniotic fluids harvested from seven-day-old chick embryos infected with type 1,type 2, or type 3 parainfluenza virus. Theamniotic fluids were concentrated 3-fold bycentrifugation and inactivated by the addi-tion of formalin to a final concentration of1:4000. The monovalent amniotic fluidharvests were then pooled to make up thefinal trivalent vaccine.

All vaccines were safety tested accord-ing to standards of the Division of Bio-logics Standards of the National Institutesof Health and each vaccine was ad-ministered to adult volunteers to demon-strate that it did not produce local or sys-temic reactions prior to its use in infants.

Composition of study group

Infants between two and seven monthsof age who attended a Child Health Cen-ter were selected for administration of vac-cine after parental consent was obtained.The infants lived at home and came from apopulation of relatively low socioeconomicstatus families, primarily Negro.

Plan of study

The schedule of vaccine administrationwas as follows: two 0.5 ml intramuscularinjections were given one month apart anda third injection was given three monthslater. Initially the parainfluenza type 1virus vaccine and the RS virus vaccinewere alternated among infants who wereotherwise comparable in age, socioeconomicstatus and location of residence (table 1).Later the type 1 parainfluenza virus vac-cine was replaced by the trivalent parain-fluenza virus vaccine in the alternation.On December 29, 1966, the injection of RSvirus vaccine was stopped for reasons tobe recounted later. From December, 1965to June, 1966, there was at all times acomparable number of infants in the RSand parainfluenza 1 vaccine groups. There-


after infants who received the trivalentparainfluenza vaccine constituted a furthercontrol group for those receiving RS virusvaccine.

Each infant under study was visited byan epidemiological nurse at least once aweek and more often if intercurrent illnessoccurred. Regardless of illness, weeklythroat and anal swab specimens were ob-tained for virus and mycoplasma studiesuntil three weeks after the third injectionof vaccine. During the immunization pe-riod serum was obtained by antecubitalvenipuncture from each infant at leastfive times, generally at monthly or bi-monthly intervals. In addition, serums wereobtained before and after the periods ofcommunity-wide RS virus prevalence. Af-ter completion of the series of vaccine in-jections, the infants were not visitedroutinely at home, but their parents wereasked to report to the vaccine study teamwhenever respiratory tract symptoms oc-curred. When ill these infante were ex-amined by an epidemiological nurse or aphysician. Throat and anal swabs andpaired specimens of blood for serum(acute and convalescent) were obtainedfor virus and mycoplasma studies.

Beginning in December of 1966 the RSvaccinees were sampled more often for thepossible presence of RS virus in the oro-pharynx than were infants who had re-ceived parainfluenza vaccine because atthat time we first suspected that exag-gerated respiratory tract illness was as-sociated with RS infection in infante whohad received RS vaccine. However, serumspecimens were obtained from all vac-cinees at two to three month intervals.These sera made it possible to compare theoccurrence of RS virus infection in the RSand parainfluenza vaccinees in an un-biased manner. Except for the more vig-orous attempts to recover RS virus fromRS vaccinees after December, 1966, allvaccinees were observed and evaluated ina similar manner. The vaccine study teamdid not influence the decision to hospitalize

TABLE 1

Composition of vaccine groups by time

Month/Year

Deo., 1966Jan., 1968Feb.MarchAprilMayJaneJalyAug.Sept.Oct.Nov.Deo.Jan., 1987Feb.Mnpnh

AprilMayJaneJaly

No. of infants receiving indicated no. of Injectionsof designated vaccine (amulatiit)

RS lot 100

1 Inj.

38

1118171921232727282931

3 Inj.

2S8

18IS16172123

Para 1 lot 23

l l n j .

27

1116182021

3 Inj.

28

1114161819

Trivalentparainfluenxa

lot 6779

l l n j .

387

10111818172021

3 Inj.

8669

101213IS19SO

an infant with infection or illness; thisdecision was made by the members of theregular hospital staff.

Laboratory procedures

Virus recovery. Anal and throat swabspecimens were collected in veal infusionbroth with 0.5 per cent bovine serum albu-min and were inoculated into tissue cultureeither immediately or after two to threehours storage at 4 C. These specimens werethen stored at — 70 C for reference and forattempts at re-isolation of recoveredagents. Two-tenths ml of each specimenwas inoculated into at least two rollertubes each of primary rhesus monkey kid-ney tissue culture and HEp-2 tissue cul-ture. The roller tube cultures were ob-served for cytopathic effect several times aweek and the monkey kidney cultures weretested for hemadsorption at seven-day in-tervals. Agents producing a cytopathic ef-fect or hemadsorption were further identi-fied by appropriate methods includingcomplement fixation for RS virus, hemad-

426 KOI ET AL.

sorption neutralization for the parainflu-enza viruses and differential hemagglutina-tion-inhibition or neutralization for theadenovi ruses.

Neutralization tests. In the early phaseof the study, neutralization tests for RSvirus were performed in HEp-2 roller tubetissue cultures, utilizing cytopathic effectas the endpoint (15). In later studies, amodification of this technique employ-ing microtiter plates was used. Serum titerswere essentially the same when tested bythe roller tube or microtiter plate tech-nique. Sixteen to 64 fifty per cent tissueculture infectious doses (TCIDBo) of theBernett strain of RS virus were used inall tests. Serum was heat inactivated (56C for 30 minutes) and diluted in 2-foldsteps starting with an initial dilution of1:2. All serum neutralization titers areexpressed as the final dilution of serum af-ter addition of an equal volume of RSvirus. The serum-virus reaction mixturealso contained a final concentration of 5per cent normal unheated guinea pigserum; this provided heat labile accessory

TABLE 2

Respiratory syncytial virus recovery from infantsand children with respiratory tract illness.

Per cent recovery by months

Month/You-

Dec, 1966Jan., 1966Feb.MarchAprilMayJuneJulyAug.Sept.Oct.Nov.Dec.Jan., 1967Feb.MarchAprilMay

No. Toted

999091

1411228563726884

10694

12313697

1059895

No. Pot.

14

1216113010202

245234770

Per cent Poe.

14.4

13.211.393.5—1.4—2.4—2.1

19.538.2356.77.1—

factor which is required to demonstrateRS antibody in certain sera (16).

Complement fixation tests. Complementfixation tests for RS virus were carriedout by a modification of the Bengstonmethod utilizing microtiter plates, 1.7 unitsof complement, and overnight incubationat 4 C (17). The antigen was preparedfrom the Long strain grown in HEp-2 cul-ture; 16 unite of CF antigen were used inall tests (2).

RESULTS

RS virus prevalence in the community.Patients with various types of acute res-piratory tract disease were studied for evi-dence of RS virus infection as part of anongoing epidemiologic investigation. In-formation from this study, which was basedupon the recovery of RS virus, provided abasis for estimating the activity of thisvirus in the community during the vaccinetrial (table 2). A period of prevalence ofRS virus infection began in January, 1966,and subsided by June, 1966. Only sporadicinfections were detected until December,1966, when another area-wide period ofprevalence began; this subsided by May,1967. Thus, there were two periods ofprevalence of RS infection during the in-terval of this investigation.

Lack of toxicity of vaccine. No unusualpain, tenderness, erythema, fever or otherimmediate or delayed local or systemicreaction was observed in infants receivingthe experimental paramyxovirus vaccines.

Antigenicity of RS vaccine. As might beexpected from their young age, all the in-fant vaccinees possessed some measurableneutralizing antibody prior to the first ad-ministration of vaccine. In most instancesthis was probably maternal antibody. RSvaccinees and a group of infante who hadreceived parainfluenza 1 vaccine during thesame time period had comparable geo-metric mean RS neutralizing antibodytiters at entry into the study (table 3).The titer rose gradually in the RS vac-cinees, from whom RS virus was not re-


TABLE 3

RS virus serum neutralizing antibody in infants who received RS virus vaccine, parainfluenza 1 virusvaccine or trivaient parainfluenza virus vaccine and from whom RS virus was not recovered during

immunization

Vaccine

RS virus vaccine (lot 100)

Parainfluenza 1 virus vaccine(lot 23)

Trivaient parainfluenza virus vac-cine (lot 6279)

No. ofinfants

23

17

6

Geometric mean RS virua neutralizing antibody titer$(reciprocal) at indicated tone:

Pre-lnject

24

16

24

1 mo. after1st inject

24

12

16

1 mo. after2nd inject.

48

12

16

prior to3rd inject*

48

8

8

3 weeki after3rd injectf

48

8

8

* 3 months after 2nd injection.t 3 weeks following 3rd injection.t <1:4 considered as 1:2 for purpose of determining geometric titer.

covered during the immunization interval,and fell in comparable parainfluenza vac-cinees. All of the RS vaccinees had de-tectable antibody at the end of the vac-cination period, and their geometric meantiter was six times that of the parainfluenzavaccinees. These findings, which are repre-sentative of the experience of the entireRS and parainfluenza type 1 vaccinegroups, are presented in lieu of the totalexperience since the serums were tested inone series of tests and thus the compara-tive results are particularly meaningful.

The response of vaccinees to RS vaccineas measured by CF and neutralization, isshown in table 4. A 4-fold or greater risein serum neutralizing antibody developedin 43 per cent of the vaccinees followingthree injections of RS virus vaccine. Bothin response to vaccine and natural infec-tion, the CF technique appeared moresensitive than the neutralization method indetecting a fourfold or greater rise in anti-body. Thirteen per cent of infants re-sponded with CF antibody after one in-jection, 62 per cent after two injections,and 91 per cent after three injections; themean fold CF antibody rise after three in-jections was 30-fold. Infants from whomRS virus was recovered prior to completionof vaccine administration were excluded

from these tabulations. However, the de-velopment of CF antibody by RS vaccineeswho later became infected with RS virusunder natural circumstances is shown inorder to compare their response with thatof infante who did not receive the RSvaccine. RS vaccinees appeared to developa more marked CF antibody response(geometric mean convalescent titer 1:384)following natural RS infection than didparainfluenza vaccinees (geometric meanconvalescent titer 1:64) following sim-ilar infection. This finding supports theview that the vaccine was antigenic or,alternately, it may be a manifestation ofthe relatively greater severity of illness inRS vaccinees than in parainfluenza vac-cinees.

The failure of RS vaccine to protectagainst infection and illness. RS virus in-fection in the RS and parainfluenza vac-cine groups could not be compared solelyfrom the virus isolation results since amore intense effort was made to recovervirus from RS vaccinees during the winterof 1966-1967. However, such a compari-son could be made from the results ofserologic studies performed with serialserums obtained before, during, and afterthe RS virus epidemics of 1966 and 1967.These serums were collected in essentially

428 HTM AL.

TABLE 4

RS complement fixing (CF) and neulralizing antibody (NA) status of infants who received inactivated RSvirus vaccine (lot 100) or inactivated parainfluenza virus vaccines (lot SS and lot 6279)

Vaccine

Respiratory syncytialvaccine (lot 100)

Parainfluenza 1 vaccineQot 23)

and

Trivalent parainfluenzavaccine Got 6279)

Time of antibodydetermination

After 1 injectionAfter 2 injectionsAfter 3 injections

Natural RS virus in-fection with recoveryof virus

After 1 injectionAfter 2 injectionsAfter 3 injections

Natural RS virus in-fection with recoveryof virus

No. ingroup

312923

16

404039

14

4-fold or greater risein antibody

CF antibody

No. infanta

4 (13%)18 (62%)21 (91%)

15 (94%)

O ( - )2 (5%)7 (15%)

13 (93%)

Meanfoldriie

2.72130

165

TO

T

O

o o

IN

71

Neutralizing antibody

No. infants

1 (3%)5 (17%)

10 (43%)

12 (75%)

N.D.*N.D.N.D.

N.D.

Meanfold rise

0.81.62.6

21

• N.D. - Not done.

the same manner from both groups and allserums were tested by a standardized CFantibody technique. Sixty-five per cent ofRS vaccinees developed serologic evidenceof RS virus infection as compared with60 per cent of infante who received para-influenza 1 vaccine, 45 per cent of thosewho received trivalent parainfluenza vac-cine, and 53 per cent of all infants whoreceived one of the parainfluenza vac-cines. Thus, the rate of infection in thevaccine groups was not remarkably dif-ferent.

Although it is clear that RS vaccine didnot provide protection against RS virusinfection, there is no evidence that itproduced an increased susceptibility toinfection. However, 80 per cent of the RSvaccinees required hospitalization at thetime of infection, whereas only 5 per centof infections among parainfluenza vac-cinees resulted in admission to the hospital(table 5). In addition, infanta who hadreceived the RS vaccine became morefleverely ill when infected with RS virus

than did the otherwise comparable infantawho received the parainfluenza vaccines.

These findings were amplified when RSand parainfluenza vaccinees were com-pared using both virus recovery and CFfor detection of infection (table 6).Twenty-three of 31 RS vaccinees experi-enced RS virus infection sometime duringor following the vaccination period. Ill-nesses among these infants included pneu-monia in six, bronchiolitis or bronchiolitiswith pneumonia in 13, and mild rhinitis,pharyngitis or bronchitis in four. Eight-een required hospitalization. Of 40 in-fants who had received one of the parain-fluenza vaccines, later RS virus infectionwas detected by virus recovery in 14 or bya fourfold or greater CF antibody rise in21—a total of 21 infections. Illnesses inthese infants included pneumonia in two,bronchiolitis or bronchiolitis with pneu-monia in two, relatively severe bronchitisor pharyngitis in two, and mild rhinitis,pharyngitis or bronchitis in 15. Only one of

B8V ILLNESS IN INFANTS DESPITE INACTIVATED VACCINE 429

RS virus infection and serious illness in comparable groups of infants receiving one or more injections ofinactivated RS and parainfluenza vaccines

Vaccine

RS lot 100

Para 1 lot 23

Trivalent parainfluenza lot6279

Total parainfluenza

Category of Infanti

At risk*RS infectionfHospitalized




No. and age of infanti daring designated time periodof RS virtu prevalence

1965-1966

No.infants

2054

202

202

Agrf(mo.)

5.1

5.0

5.0

1966-1967

No.infants

25t1512

m101

2090

37191

Age|(mo.)

12.7

15.8

8.4

11.8

Total No.infants

3120 (65%) I)16(80%)V

2012 (60%) HK8%)H

209 (45%) 00

4021 (53%)||

1 (5%)U

* No prior natural infection.t As indicated by 4-fold or greater CF antibody rise only; reinfection counted only once.X One infant was not available to follow up.§ Mean age at peak of RS prevalence.|| Per cent of infants at risk who sustained RS infection.U Per cent of infected individuals who were hospitalized.

these infante required hospitalization atthe time of RS infection.

The severity of illness in the RS vac-cinees was greater than in unvaccinatedinfante and children who were admitted tothe hospital during the community wideoutbreak of RS virus infection. The meanperiod of hospitalization of the RS vac-cinees was 10.5 days, whereas the meanperiod of 30 age-matched unvaccinated in-fante with RS bronchiolitis and/or pneu-monia was 6.7 days.

The RS vaccinees who became seriouslyill during RS infection included both re-cent vaccinees and individuals whosecourse of injections was completed as longas 11 months before the onset of illness(table 7). Those who were infected andbecame ill included two infante who hadreceived only one injection and eight whohad received two injections. All of the in-

fante possessed moderate to high levels ofserum neutralizing antibody just prior toand also during the acute phase of illness.Approximately a third of those who becameinfected had RS virus CF antibody intheir serum prior to illness or during theacute phase of illness; however, few ofthese sera had CF antibody levels com-parable to those seen during convalescencefrom RS virus illness. These findings in-dicate that vaccine-induced serum neu-tralizing antibody persisted until the timeof infection. In addition, the findingsclearly indicate that moderate to highlevels of serum neutralizing antibody, assuch, did not provide effective protectionagainst RS virus lower respiratory tractdisease.

Two infante died, one at age 14 months,the other at age 16 months. Each had re-ceived three inoculations, one beginning at

430 1TTM

TABLB 6

RS virus infection and iUness in groups of infants after receiving 1 or more injections of inactivated respira-tory syncyiial and parainfluenza vaccines

Vaccine received No.vacdneei

No. vaccineei with later RSinfection as indicated by

Viruirecovery

CFantibody

riie

Viruirecoveryand/orCFriM

No. vacrinees with designated illnenat time of RS infection

BforB-P BR-PHJ URI|

No.requiring

hospitaKza-tion at time

of RS i jinfection

RS lot 100Para 1 lot 23Trivalent parainflu-

enza lot 6279Total parainfluenza

312020

40

1877

14

20129

21

23129

21

620

1302

4106

15

1810

* P = Pneumonia.t B •=• Bronchiolitis.t BR-PH = Severe bronchitis-pharyngitis.§ URI ™ Mild rhinitis, pharyngitis and/or bronchitis.

age two months, the other at age fivemonths. The major findings at post mortemexamination were extensive bronchopneu-monia and patchy atelectasis with emphy-sema and pneumothorax in one infant.Microscopic histologic findings indicatedperibronchiolar monocytic infiltration withsome excess in eosinophiles, findings whichare consistent with the limited literature onthe histologic appearance of bronchiolitis.RS virus was grown readily from therespiratory tract and lung in both infants.At least 104 TCID60 of RS virus werepresent per gram of lung of the younger in-fant.

In addition a pure culture of E. coli wasrecovered from the trachea, blood, spleenand lung of one infant and a pure cultureof KlebsieUa was recovered from the lung,trachea and nose culture of the other.

The age at which RS vaccinees developedRS virus lower respiratory tract disease is6hown in table 8. This age distributiondiffers from that of unvaccinated infantswho developed RS virus disease during theJanuary, 1966-June, 1966, and December,1966-April, 1967, epidemics. In the com-munity-wide outbreak the occurrence ofsevere illness requiring hospitalization washighest during the first two months of lifeand decreased significantly with increasing

age. The RS vaccinees could not be com-pared, in a strict sense, with the un-vaccinated infant population because vac-cine was not administered until the vac-cinees were two to seven months of age.However, the frequent occurrence ofserious RS virus respiratory tract diseasein vaccinees who were over six months ofage suggested that prior administration ofvaccine produced an alteration in host re-sponse wherein the older vaccinees de-veloped the type of lower respiratory tractdisease most commonly associated withinfection of early infancy.

DISCUSSION

From these findings it seems clear thatinfants who received a monkey kidney tis-sue culture grown, inactivated, 100X con-centrated, alum precipitated respiratorysyncytial virus vaccine not only were notprotected against infection, but para-doxically they experienced an altered, ex-aggerated clinical response when naturalRS infection occurred. These findings wereentirely unexpected. We now feel that insome way administration of the inactivatedRS vaccine was responsible for the ex-aggerated illness caused by RS virus in-fection. We lack a definite explanation forthis phenomenon, except to say that the

TA

BL

E 7

Infa

nts

rece

ivin

g in

acti

vate

d re

spir

ator

y sy

ncyl

ial v

irus

vac

cine

lot

100

who

lat

er u

nder

wen

t na

tura

l R

S vi

rus

infe

ctio

n

Vac

dnee

No. 1 2 3 4 5 6 7 8

ft

9 811 10 11 12 13 14 16 16 17 18§

19

"20 21

Dia

gnos

is

B B-P

JU

RI»

P P P P B*t

B B-P

B-P

B B P B B-P

JB

-PB B U

RI*

P B-P

Age

/Mon

ths

Fir

st l

noc.

2 2 2 2 2 2i 3 3 3 3 3 3 3 3* 5 6 5 6 6 6 6 7

7 14 11 12 3* 12 5 5 5 15 4 9i 4 13 6 16 18 8 16 17 11 13

Evi

denc

e of

RS

infe

ctio

n

Vir

uire

cove

red

+ + + + + + 0 + + + + + + + + + + 0N

.T.

+ + +

CF

ant

i-bo

dy r

ise

+N

.T.

0 + + + + + + + + + + + +N

.T.

+ + + + + +

No.

of

inoc

.be

fore

i Une

u

3 3 3 3 1 3 2 2 2 2 2 3 1 3 2 3 3 2 3 3 3 2

Tim

e af

ter

last

in

octo

illn

ess

28

d8^

m4

m6

m52

d 5 H

m2

3d

38

d30

dll

m14

d 2 m

lm 5 m

17 d 5M

m9

m36

d 6 m

8 m

1 m

4m

Day

s be

-tw

een

last

seru

m &

acut

e ill

-ne

ss s

erum

29 31 70 11 28 32 28 8 30 32 15 5 6 13 18 8 17 6 16 51 8 26

Rec

ipro

cal

of R

S vi

rus

CF

ant

ibod

y ti

tere

Pre

inje

ct.

8<

4<

4<

4<

4<

4<

4<

4<

4<

4 4<

4 4<

4<

4<

4<

4<

4<

4<

4<

4<

4

Las

t se

-ru

m b

efor

eill

ness

<4

<4 64 4

<4

<4

<4 16 <4 16 <4 8

<4

<4

<4

128

<4 64 16 4 8 16

Acu

teill

ness

seru

m

<4

<4

—4

<4

128 32 —8

<4 16 <4

<4 32 64 <4

— — <4 32 16

Con

vale

-sc

ent

leru

m 256

—32 256 32

>512 12

812

825

651

2>5

12>5

12 32>5

12>5

12— 51

251

2>5

12>5

12>5

12>1

28

Rec

ipro

cal

of R

S vi

rus

neut

raliz

ing

anti

body

lit

ers

Pre

inje

ct

64 12 128 16 48 24 32 24 12 12 48 48 256 48 24 16 6 6 12 6 96 8

Las

t se

-ru

m b

efor

eill

ness

48 12 64 4 24 48 32 12 12 24 32 24 128

128 12 48 12 32 96 16 192 32

Acu

teill

ness

seru

m

64 4 — 96 24 128 48 — — 24 24 16 96 96 64 — 16 — — 12 384 48

Con

vale

scen

tse

rum

192 — 48

N.T

.96 384 64 64 48 768

or >

768

or >

384

768

or >

384

768

or >

— 128

384

768

or >

16 384

384

* N

ot

adm

itted

to

ho

spita

l. f

Influ

enza

B

als

o re

cove

red.

t

Dec

ease

d.

§ H

ospi

taliz

ed

at

othe

r ho

spita

l.D

ecem

ber

1966

out

brea

k.

** R

einf

ectio

n w

ith i

nter

val

of 2

8 da

ys. O

n bo

th o

ccas

ions

infa

nt h

ad U

RI

with

low

gra

de f

ever

.=

day

(s).

m

= m

onth

(s).

Rei

nfec

tion

occu

rred

du

ring

B =

Bro

nchi

oliti

a.Pn

eum

onia

. N

.T.

= N

ot t

este

d.

432 KIM ET AL.

TABLE 8

Age distribution of RS virus illness in vaccinegroup compared with thai of community

Age at which RSbronchiolitii or

pneumonia occurred

1-23-46-67-89-10

11-1213-1416-1617-1819-2021-2223-24

Total

Patients fromrnmintnii ty

No.

3621181354452341

116

Per cent oftotal

31]18 65%16J1143342331

RS vaccineaNo.

34213321

•63%

19

vaccine in some way altered the host re-sponse to natural RS infection. This phe-nomenon is presumably similar to the"paradoxical" vaccine effect which hasbeen reported with parenterally admin-istered, rickettsial vaccines, trachoma vac-cine, Mycoplasma jmeumoniae vaccine andmore recently with inactivated measlesvaccine (18-22).

In retrospect the above findings recallexperience with a group of 54 infante whoreceived a 4X concentrated, formalin-in-activated, alum precipitated RS virus vac-cine in 1962 and 1963. In those infante weconcluded that the vaccine was not suffi-ciently antigenic to be considered for fur-ther trials. In addition we concluded thatthe test vaccine was completely ineffectivein preventing illness or infection because21 of the 54 infants who received two orthree injections experienced natural RSvirus infection during the subsequent pe-riod of high prevalence. Ten of these infec-tions were associated with severe lowerrespiratory tract illness requiring hos-pitalization. Three of these hospitalized in-fante had 6erum neutralizing antibodylevels at the time of illness which may have

reflected either persistence of maternalantibody or an immunogenic effect of thepoorly antigenic RS virus vaccine. At thetime we felt that the high incidence of in-fection and illness in vaccinees was pos-sibly outside the range of normal; how-ever, there was no concurrent controlgroup with which to compare the unusualincidence of illness in the vaccinees. Inview of the recent findings, it is likely thatthe high incidence of serious illness in theinfante who received the earlier poorlyimmunogenic RS virus vaccine was as-sociated with administration of the vac-cine.

The parainfiuenza type 1 vaccine, whichwas given to alternate vaccinees during thefirst half of the investigation, was pre-pared in a manner virtually identical tothat of the RS vaccine. Both vaccines wereprepared from virus grown in vervet mon-key kidney tissue cultures. Both viruseswere concentrated 100-fold and in both in-stances alum was used as an adjuvant.Since the parainfluenza type 1 vaccineesdid not exhibit an apparent altered re-sponse to RS virus infection, it is unlikelythat monkey kidney antigens, either insedimentable form unassociated with virusor incorporated into the virion envelope,were responsible for the exaggerated clin-ical response of RS vaccinees. Thus, acomponent of RS virus itself, and not atissue culture host cell antigen, probablywas responsible for inducing an alteredstate of reactivity to RS infection in RSvaccinees.

In reviewing the incidence of RS viruslower respiratory tract illness at Children'sHospital, Washington, D. C. from 1959 to1966, we observed that the highest rate ofsuch serious disease occurred in the firstfour months of life, at a time whenmaternally transmitted antibody was pres-ent at a moderate to high level (13). Thispattern was repeated during the 1966-1967RS virus epidemic in Washington. Thesefindings, in addition to indicating thatserum neutralizing antibody by itself does


not provide effective protection, suggest tous that passively acquired antibody mayplay a role in the pathogenesis of RSvirus lower respiratory tract disease ofearly life (23). Possibly antigen-antibodycomplexes at the respiratory epithelial sur-face initiate a sequence of events involv-ing complement fixation, chemotaxis andleukocyte damage which leads to thebronchiolar pathology seen in serious RSvirus disease (24, 25).

Recent studies with another paramyxo-virus, type 1 parainfluenza virus, indicatedthat local respiratory tract secretory anti-body was more important in resistance toinfection than was serum antibody (26).If a similar situation obtains for RS virusand if serum antibody plays a role in RSvirus pathogenesis, one can then postulatethat parenterally administered inactivatedRS vaccine produced its alteration of hostresponse by increasing both the level andpersistance of serum antibody in a hostwho lacked respiratory tract secretoryantibody. In this sense inactivated vac-cine may have acted to extend the periodof vulnerability of vaccinees beyond thefirst few months of life.

REFERENCES

1. Chanock, R. M., Kim, H. W., Vargosko, A. J.,DeLeva, A., Johnson, K. M., dimming, C.and Parrott, R. H. Respiratory syncytialvirus. I. Vims recovery and other observa-tions during 1960 outbreak of bronchiolitis,pneumonia, and other minor respiratory dis-eases in children. JAMA., 1961, 176: 647-653.

2. Parrott, R. H., Vargosko, A. J., Kim, H. W.,Cumming, C , Turner, H., Huebner, R. J.and Chanock, R. M. Respiratory syncytialvirus. II. Serologic studies over a 34-monthperiod of children with bronchiolitis, pneu-monia, and minor respiratory diseases.JAMAn 1961,176: 653-657.

3. Chanock, R. M., Parrott, R. H , Vargosko, A.J., Kapikian, A. Z., Knight, V., and Johnson,K. M. IV. Respiratory syncytial virus. Amer.J. Pub. Health., 1962,6£: 918-925.

4. Chanock, R. M. and Parrott, R. H. Acuterespiratory disease in infancy and childhood:present understanding and prospecta forprevention. Pediatrics, 1965, S8: 21-39.

5. Beem, M., Wright, F. H., Hamre, D., Egerer,R. and Oehme, M. Association of the chim-panzee coryza agent with acute respiratorydisease in children. New Eng. J. Med., 1960,S63: 523-530.

6. Beem, M., Wright, F. H., Fasan, D. M., Egerer,R. and Oehme, M. Observations on theetiology of acute bronchiolitis in infants. J.Pediat., 1962,61: 864-869.

7. Berglund, B., Vihma, L. and Wickstrom, J.Respiratory syncytial virus studies on chil-dren hospitalised during an outbreak ofrespiratory illness in Finland. Amer. J.Epid., 1965,81: 271-281.

8. Hamparian, V. V., Ketler, A., Hilleman, M. R.,Reilly, C. M , McClelland, I., Comfeld, D.and Stokes, J., Jr. Studies of acute respira-tory illnesses caused by respiratory syncytialvirus. I. Laboratory findings in 109 cases.Proc. Soc. Exper. Biol. Med., 1961, 106:717-722.

9. Holzel, A., Parker, L., Patterson, W. H., White,L. L. R., Thompson, K. M. and Tobin, J.OIL. The isolation of respiratory syncytialvirus from children with acute respiratorydisease. Lancet, 1963,1: 295-300.

10. Lewis, F. A., Rae, M. L., Lehmann, N. I., andFerris, A. A. A syncytial virus associatedwith epidemic disease of the lower respira-tory tract in infanta and young children.Med. J. Aust., 1961, 2: 932-933.

11. Sterner, G., Wolontis, S., Bloth, B., anddeHevesy, G. Respiratory syncytial virus.An outbreak of acute respiratory illnesses ina home for infants. Acta Paediat. Scand.,1966, 65: 273-279.

12. Tyrrell, D. A. J. (secy.). A collaborative studyof the aetiology of acute respiratory infec-tions in Britain, 1961-1964. Medical ResearchCouncil Working Party on Acute RespiratoryVirus Infections. Brit. Med. J., 1965, 2: 319-326.

13. Chanock, R. M., Smith, C. B., Friedewald, W.T., Parrott, R. H., Forsyth, B. R, Coates,H. V., Kapikian, A. Z. and Gharpure, M. A.Resistance to parainfluenza and respiratorysyncytial virus infection—implications foreffective immunisation and preliminary studyof an attenuated strain of respiratory syn-cytial virus. In First International Conferenceon Vaccines Against Viral and RickettsialDiseases of Man. Washington, D.C., PanAmerican Health Organization, 1967, Sci.Pub. 147, pp. 53-61.

14. Biological Producta—Public Health ServiceRegulations, Title 42, Part 73, Dep't. HEW,U.S.P.H3. Pub. No. 437, revised Jan., 1967.

15. Chanock, R. M., Roizman, B. and Myers, R.Recovery from infants with respiratory ill-

434 TZTM E T A L .

ness of a vims related to chimpanzee coryzaagent (CCA). 1. Isolation, properties andcharacterization. Amer. J. Hyg., 1957, 68:281-290.

16. Potash, L., Lees, R. S. and Sweet, B. H. In-creased sensitivity for detection of respira-tory syncytial virus neutralizing antibodyby use of unheated sera. Bact. Proc, 1966,p. 122.

17. Sever, J. L. Application of a microtechniqueto viral serological investigators. J. Immunol.,1962,88: 320-329.

18. Cox, H. R., Greenberg, B. G, Kleinman, H.and Meier, P. The present status of poliovaccines. 111. Med. J , 1960,118: 160-168.

19. Grayston, J. T , Woodbridge, R. L. and Wang,S. Trachoma vaccine studies on Taiwan.Ann. N.Y. Acad. Sci, 1962, 98: 362-367.

20. Chanock, R. M., Smith, C. B , Friedewald, W.T , Gutekunst, R., Steinberg, P., Fuld, 8.,Jensen, K. E., Senterfit, L. B. and Prescott,B. Mycoplasma pneumoniae infection: pros-pects for live and attenuated vaccines. InFirst International Conference on VaccinesAgainst Viral and Rickettsial Diseases ofMan. Washington, D.C, Pan American

Health Organisation, 1967, Sci. Pub. 147,pp. 132-140.

21. Fulginiti, V. A, Eller, J. J., Downie, A. W.,and Kempe, C. H. Altered reactivity tomeasles virus. Atypical measles in childrenpreviously immunised. JA.MA. SOS: 101-106.

22. Buser, F. Side reaction to measles vaccinationsuggesting the arthus phenomenon. New Eng.J.Med., 1967, S77: 250.

23. Chanock, R. M., Parrott, R. H., Kapikian, A.Z., Kim, H. W, and Brandt, C. D. Possiblerole of immunologic factors in pathogenesisof RS virus lower respiratory tract disease.Perspectives in Virology, Vol. VI. In press.

24. Ward, P. A., Cochrane, C. G., and Mueller-Eberhard, H. J. The role of serum comple-ment in chemotflTJH of leukocytes in vitro.J. Exp. Med., 1965,1SS: 327-346.

25. Rapp, H. Jn and Borsos, T. Complement re-search, fundamental and applied. J.A.MA.,1966,198: 1347-1354.

26. Smith, C. B , Purcell, R. H., Bellanti, J. A.,and Chanock, R. M. Protective effect ofantibody to parainfluenza type 1 virus. NewEng. J. Med, 1966, £75: 1146-1152.

Documents

RESPIRATORY SYNCYTIAL VIRUS DISEASE IN INFANTS …