S215

Combined Immunization of Infants with Oral and Inactivated Poliovirus Vaccines: Results of a Randomized Trial in The Gambia, Oman, and Thailand

WHO Collaborative Study Group on Oral and World Health Organization, Geneva, Switzerland Inactivated Poliovirus Vaccines*

To assess an immunization schedule combining oral (OPV) and inactivated poliovirus vaccines (lPV), a clinical trial was conducted in The Gambia, Oman, and Thailand. Children were randomized to receive OPV at birth and at 6, 10, and 14 weeks of age; OPV at birth followed by both OPV and Downloaded from https://academic.oup.com/jid/article/175/Supplement_1/S215/878782 by guest on 24 September 2021 IPV at 6, 10, and 14 weeks of age; or placebo at birth followed by IPV at 6, 10, and 14 weeks of age. Serum specimens were available at 24 weeks for 1291 (77%) of 1685 enrolled infants. In the combined-schedule group, the proportion of children seropositive at 24 weeks was 95%-99% for type 1,99%-100% for type 2, and 97%-100% for type 3. In The Gambia and Oman, the combined schedule performed significantly better than OPV for type 1 (95%-97% vs. 88%-90%) and type 3 (97%-99% vs. 72%-73%). Across the study sites, IPV given at 6, 10, and 14 weeks of age provided inadequate protection against poliovirus. The combined schedule provided the highest levels of serum antibody response, with mucosal immunity equivalent to that produced by OPV alone.

In 1988, the World Health Assembly established the target which polio is endemic are supplemental OPV delivery strate­ of global eradication of poliomyelitis by the year 2000 [1]. gies, including national immunization days and house-to-house Since then, remarkable progress has occurred in freeing many immunization in high-risk areas, strategies that were successful countries from poliomyelitis [2, 3]. Extensive use of trivalent in the elimination effort in the Americas and are being adopted oral poliovirus vaccine (OPV) has been associated with the in other parts of the world. interruption of wild poliovirus circulation in the Americas, Nonetheless, in developing countries there is wide variation where the last case of paralytic disease associated with wild in the serologic response ofchildren to OPV, with overall levels poliovirus isolation occurred in August 1991 [4]. ofseroresponse below those seen in industrialized countries. A Current poliomyelitis eradication strategies recommended by review of 32 studies in developing countries found that after the World Health Organization (WHO) focus on the early and three doses ofOPV, the mean proportion of infants with detect­ intensive use ofOPV with routine delivery of doses at birth and able levels of serum neutralizing antibody was only 73% 6, 10, and 14 weeks of age. Also recommended for countries in (range, 36%-99%) for type 1,90% (range, 71%-100%) for type 2, and 70% (range, 40%-99%) for type 3 [5]. A recent large-scale randomized trial in Brazil and The Gambia has confirmed these findings [6]. Even after eight OPV doses deliv­ Published: WHO Collaborative Study Group on Oral and Inactivated Polio­ ered in mass campaigns, gaps in immunity, as defined by serum virus Vaccines. Combined immunization of infants with oral and inactivated antibody levels, persist in some countries, especially for type poliovirus vaccines: results of a randomized trial in The Gambia, Oman, and 3 [7]. There is also evidence that OPV may not always succeed Thailand. Bull World Health Organ 1996;74:253-68. Voluntary and informed consent for participation of children in this study in preventing transmission of wild poliovirus, even when vac­ was obtained from the parent or guardian in accordance with ethical principles, cine coverage is excellent. Outbreaks of poliomyelitis have including the provisions of the Declaration of Helsinki, and the additional occurred in some countries where coverage with three or more requirements of local and national authorities. The study was approved by the Ministry of Public Health, The Gambia; the Ethical Committee of the Medical doses of OPV has been high, notably in Brazil [8], Bulgaria Research Council Laboratories, The Gambia; the Ministry of Health, Oman; [9], The Gambia [10], Jordan [11], Israel [12], Malaysia [13, the National Maternal Child Health Committee, Oman; the Maternal Child 14], Namibia [15], Oman [16, 17], and Saudi Arabia [18]. In Health Committee, North Batinah Region, Oman; the Ministry of Public Health, Thailand; the Ethical Committee, Phrarnongkutklao College of Medi­ these settings, vaccine efficacy has appeared to correlate with cine, Thailand; the Institutional Review Board, Centers for Disease Control seroconversion rates, and some children with well-documented and Prevention, Atlanta; and the Secretariat Committee for Research Involving immunization histories have contracted paralytic disease. Human Subjects, World Health Organization, Geneva. Financial support: WHO Global Programme for Vaccines and Immunization, Some industrialized countries have become polio-free using Geneva, using funds donated by the governments of Finland and Sweden, the inactivated poliovirus vaccine (IPV), but this vaccine has not Rockefeller Foundation, and the United Nations Development Programme. been recommended by WHO for polio eradication in devel­ Reprints, correspondence, and copies of the full study report (document WHONRD/GEN/95.1): Dr. Susan E. Robertson, Vaccine Research and Devel­ oping countries. When given to children in developing coun­ opment, Global Programme for Vaccines and Immunization, World Health tries at intervals of ;?:2 months, starting at ;?:8 weeks of age, Organization, 1211 Geneva 27, Switzerland. IPV has evoked excellent serologic response, but it is not * Study group members are listed after the text. known how long this protection lasts [19]. Because IPV does The Journal oflnfectious Diseases 1997;175(Suppll):S215-27 not produce the high levels of intestinal mucosal immunity seen © 1997 by The University of Chicago. All rights reserved. 0022-1899/97/7581":"0038$01.00 with OPV, a person who is protected from paralytic disease by S216 WHO Collaborative Study Group JID 1997;175 (Suppl1)

serum antibody may still excrete and transmit poliovirus to Table 2. Vaccine doses by age in 3 study groups in a randomized others. Moreover, unlike OPV, IPV is a killed vaccine, and it trial in The Gambia, Oman, and Thailand. cannot, spread secondarily to contacts. OPV and IPV A few countries or areas have become polio-free using a Age OPV group group IPV group combination of OPV and IPV, notably Denmark [20] and the Palestine Self-Rule Areas of Gaza and West Bank [21]. How­ Birth* OPV OPV Placebo ever, these countries or areas use relatively complex schedules 6 weeks OPV + DTP OPV + DTP-IPV Placebo + DTP-IPV 10 weeks OPV DTP OPV + DTP-IPV Placebo DTP-IPV that are not completed until the first birthday or older (table + + 14 weeks OPV + DTP OPV + DTP-IPV Placebo + DTP-IPV

1). In the World Health Assembly requested research to Downloaded from https://academic.oup.com/jid/article/175/Supplement_1/S215/878782 by guest on 24 September 2021 1988, 24 weeks Challenge Challenge Challenge develop additional polio immunization strategies that could speed the success of polio eradication [1]. Subsequently, the NOTE. DTP = diphtheria-tetanus toxoids-pertussis vaccine; DTP-IPV = Global Advisory Group of the WHO Expanded Programme combined DTP and IPV given as single injection. * At birth or first contact after birth. on Immunization specifically requested the assessment of a combined polio immunization schedule that would be practical to implement in developing countries, because it was antici­ pated that such a schedule could provide infants with the bene­ In The Gambia, field work was done between January 1990 and February 1991. A total of 505 subjects were recruited from the ficial effects of both poliovirus vaccines at an early age [22]. weekly immunization clinic at the Health Center in Brikama, a In response to these recommendations, the WHO Collabora­ town of 10,000 population. Although there was no active surveil­ tive Study Group on Oral and Inactivated Poliovirus Vaccines lance for paralytic poliomyelitis in The Gambia, 1 case was re­ was formed. Here we report the results ofa randomized clinical ported in 1990 and 1 case in 1991 [23]. Neither of these patients trial of a combined polio immunization schedule in The Gam­ lived in Brikama. bia, Oman, and Thailand. To provide findings broadly applica­ In Oman, field work was done between October 1990 and April ble to future policy recommendations, the combined schedule 1992. A total of 630 subjects were recruited from infants born at was studied in different regions of the world. The study group Sohar Hospital, Sohar, whose families were residents of one of thought it was important to study a simple schedule with simul­ the 27 villages of Sohar District (total population, 65,535). Since taneous delivery of OPV and IPV at ages routinely recom­ 1989, Oman has conducted intensive surveillance for poliomyelitis, mended by WHO for immunization in developing countries, with investigation of all cases of acute flaccid paralysis reported with polio immunization completed before 4 months of age among children < 15 years of age [17]. No cases due to wild (table 1). The trial assessed both the serum immune response poliovirus were detected in 1990 or 1992 [23]. In 1991, 4 cases and the mucosal immune response (as judged by protection on due to wild type 3 virus were confirmed, all in persons who lived outside the study district. In response, OPV mass campaigns were challenge with monovalent type 1 OPV at 6 months of age). conducted from April to December 1991 in parts ofOman adjacent Methods to, but not in, the study district. After completion ofthe study, the Study subjects. The study enrolled 1685 infants, who were free OPV mass campaigns were extended into Sohar District. of apparent illness at their initial examination. In Thailand, field work was done between June 1991 and July 1992. A total of 550 subjects were recruited from infants born at Phramongkutklao Hospital, Bangkok. The hospital catchment area Table 1. Combined polio immunization schedules used by Denmark includes the entire city of Bangkok (total population, 8 million). and the Palestine Self-Rule Areas of the West Bank and Gaza, com­ Thailand initiated intensive poliomyelitis surveillance in 1991. pared with the schedule used by the present study in The Gambia, There were 5 cases of paralytic poliomyelitis due to wild virus Oman, and Thailand. reported from Thailand in 1991 and 9 in 1992 [23]. Most ofthese patients lived in the northeastern part ofthe country; no cases were Age Present reported from Bangkok. (months) Denmark West Bank Gaza study The study sites in The Gambia and Oman consisted of rural Birth OPV populations of low socioeconomic status, mostly illiterate, with 1 M1-0PV OPV + IPV large extended families. Pit latrines were common in both sites. 2 OPV OPV + IPV OPV + IPV In The Gambia, water was obtained from community wells, 3 OPV + IPV OPV + IPV whereas each house in the study district in Oman had a tank of 4 OPV + IPV chlorinated water. The study site in Bangkok enrolled families of 5 IPV OPV + IPV OPV middle socioeconomic status, who were well-educated; most had 6 IPV OPV only one child and lived in areas with modem sanitation. 12 OPV OPV Study design. Each enrolled child was randomly assigned to 15 IPV a study group using a computer-generated list. The groups received 24 OPV 36 OPV either OPV at birth and 6, 10, and 14 weeks of age; OPV at birth 48 OPV followed by both OPV and IPV at 6, 10, and 14 weeks of age; or placebo at birth followed by IPV at 6, 10, and 14 weeks of age NOTE. M1-0PV = monovalent type 1 OPV. (table 2). Each O.I-mL dose of OPV or placebo was administered JID 1997; 175 (Suppl 1) Infant Immunization with OPV and IPV S217

as 2 drops into the back of the mouth with the dropper supplied D antigen units of type 2 (MEF1), and 32 D antigen units of type by the manufacturer. Children in the DPV-only group received 3 (Saukett) poliovirus strain. (This formulation has been termed diphtheria-tetanus toxoids-pertussis (DTP) vaccine at 6, 10, and enhanced-potency IPV, eIPV, and E-IPV in other publications.) 14 weeks of age, while children in the IPV-only and combined­ The IPV was combined with alum-adjuvanted DTP vaccine and schedule groups received DTP-IPV as a combined vaccine at the stabilized with 2-phenoxyethanol. Two lots ofIPV were used: one same ages (table 2). DTP or DTP-IPV was administered as a 0.5­ lot in The Gambia throughout the study and in Oman until 4 mL dose by intramuscular injection in the lateral thigh. In Oman August 1991, the second lot in Oman from 5 August 1991 and in and Thailand, vaccines were color-coded and investigators and Thailand throughout the study. DTP-IPV and DTP vaccines participants remained blinded to the vaccine administered; in The (Pasteur-Merieux Serums et Vaccins) were supplied in single-dose

Gambia, investigators were not blinded to the identity of study prefilled syringes. DTP-IPV and DTP were shipped from the manu­ Downloaded from https://academic.oup.com/jid/article/175/Supplement_1/S215/878782 by guest on 24 September 2021 vaccines. facturer to the study site at 2-8°C with vaccine cold-chain monitor To provide an indirect measure of mucosal protection against cards (to detect exposure to heat) and freeze-watch monitors (to type 1 poliovirus, infants were challenged at 24 weeks of age with detect freezing) [25]. These vaccines were stored at 2-8°C with 6 a dose of 10 TCIDso of monovalent type 1 OPV. The design for both types of temperature monitors. A single cold-chain break the challenge was based on pilot studies done during 1989-1990 occurred in Thailand after a weekend thunderstorm, when the tem­ in The Gambia [24]. Excretion of type 1 poliovirus at 7 days perature of the study vaccine refrigerator reached 14°C possibly after challenge was interpreted as a nonprotective level ofmucosal for as long as 48 h. The manufacturer and independent experts immunity. advised that the temperature exposure reported in this cold-chain Additional vaccines were administered during the study period, break would not have been sufficient to affect the potency of the as routinely recommended in each country. In all three countries, IPV. infants received a dose of bacille Calmette-Guerin vaccine at the Potency of the IPV was tested on the bulk product before it had same visit as the birth dose of OPV or placebo. In The Gambia been mixed with DTP vaccine, since after mixing, it is not possible and Oman, hepatitis B vaccine was given simultaneously with to quantitatively test the potency ofthe IPV in the combined prod­ some study vaccine doses. In all study sites, infant multivitamin uct [26]. Tests conducted by the manufacturer showed that the preparation was offered to parents, but it was not administered at first IPV lot had 38, 10, and 33 D antigen units ofpoliovirus types the study visits. In Thailand, infant fluoride preparation was offered 1, 2, and 3 per dose. The second IPV lot had 42, 9, and 34 D to parents, but it was not administered at the study visits. antigen units of types 1, 2, and 3 per dose. These results were Vaccines. Trivalent OPV (SmithKline Beecham Biologicals, confirmed by laboratories at two WHO Collaborating Centers. 6 s.s Rixensart, Belgium) was formulated to contain 10 , lOS, and lO Review of the IPV potency data by an independent expert con­ TCIDso ofpoliovirus types 1,2, and 3 per O.I-mL dose, represent­ cluded that the two IPV lots did not have different potencies. ing a 10:1:3 ratio of the virus types. The vaccine was stabilized Specimen collection. To assess serum neutralizing antibody with magnesium chloride and was the same product routinely sup­ titers against poliovirus types 1, 2, and 3, at all sites a I-mL plied through UNICEF at the time the study started in January blood specimen was obtained from each infant by fingerstick or 1990. OneIot of trivalent OPV was used in The Gambia and venipuncture at 14 and 24 weeks of age. In Oman and Thailand, Oman, and a second lot was used in Thailand. A single lot of cord blood specimens were collected to determine maternally de­ monovalent type 1 OPV (SmithKline Beecham Biologicals) con­ rived antibody. In Oman, additional blood specimens were ob­ 6 taining 10 TCIDso of poliovirus type 1 was used for the challenge tained at 6 and 10 weeks of age. The serum was separated, stored study at all three sites. These vaccines were shipped on dry ice at - 20°C, and shipped by air on dry ice to the Enterovirus Labora­ from the manufacturer to the study sites. Vaccine cold-chain moni­ tory, Centers for Disease Control and Prevention (CDC), Atlanta, tor cards were activated by the manufacturer before shipment and for testing. remained with the vaccine during shipment and storage [25]. Vac­ Two stool specimens were collected: the firstjustbefore delivery cines were stored at - 20°C, with temperature monitoring twice of the challenge with type 1 OPV at 24 weeks of age and the daily. second 7 days after challenge. In The Gambia and Oman, stools Potency tests at two WHO Collaborating Centers showed mean were collected by parents in stool containers and transported to potency for the first DPV lot to be 6.52, 5.50, and 5.93 log TCIDso the laboratory the same day. Once received, stool specimens were per dose for poliovirus types 1, 2, and 3, respectively. Mean po­ frozen pending virus isolation studies. In Thailand, rectal stool tency for the second OPV lot was 6.49, 5.25, and 5.68 log TCIDso samples were collected by study physicians with a thin plastic tube per dose for types 1, 2, and 3, respectively. For both OPV lots, [27]. The Thai stool specimens were transported in a cold box to there was a <0.5 log TCIDso drop in the titer in the stability test the laboratory, where virus isolation studies were begun the same (48 h at 37°C) and in field samples of DPV randomly collected day. from each of the study sites. Mean potency of the challenge OPV Serology. Each serum specimen was tested for all three polio­ was 6.24 log TCIDso of type 1 virus per dose, with a <0.5 log virus types in triplicate using a modified microneutralization assay TCIDso drop in the titer in the stability test and in field samples. developed by the Enterovirus Laboratory, CDC. This assay has DPV placebo administered to the IPV group at each visit con­ been used at CDC to determine poliovirus antibody titers through­ sisted of sterile water in The Gambia and molar magnesium chlo­ out the developing world [6, 8, 16, 28, 29]. Serial dilutions of ride solution (the same composition as the stabilizer of the OPV) serum (starting at 1:8 and ending at 1:1024) were incubated with in Oman and Thailand. 100 TCIDso of poliovirus types 1, 2, and 3 at 36°C for 3 h before IPV (Pasteur-Merieux Serums et Vaccins, Lyon, France) was 1-2 X 104 HEp-2 (Cincinnati) cells were added to each well. The formulated to contain 40 D antigen units of type 1 (Mahoney), 8 HEp-2 (Cincinnati) cell line is particularly sensitive for poliovi- S218 WHO Collaborative Study Group JID 1997;175 (Suppl1) ruses. Before December 1991, sera from The Gambia were tested statistically significantdifferences among study groups at each site using the same method, except that HeLa cells were used; the regarding baseline characteristics, seroprevalence rates, serocon­ transition from HeLa to HEp-2 (Cincinnati) cells was carefully version rates, GMTs, and virus excretion rates. calibrated. The CDC laboratory participated in an international collaborative study of poliovirus neutralizing antibody tests and results were comparable with those from other laboratories [30]. Serologic results were reported as reciprocal titers of the serum Results dilutiondemonstrating 50% inhibition.Seroprevalencewas defined by the presence of any detectable antibody (titer ;::8). Seropreva­ Enrollment and follow-up. A total of 1685 infants (505 in lence estimates were based on sera from all infants who had re­ The Gambia, 630 in Oman, 550 in Thailand) were enrolled. Downloaded from https://academic.oup.com/jid/article/175/Supplement_1/S215/878782 by guest on 24 September 2021 ceived the vaccines with intervals between doses that fell within Within each country, there were no significant differences be­ a range of22-9l days. Seroconversionwas defined as the presence tween study groups in median or mean age at study visits, sex, of any detectable antibody (titer ;::8) in an infant who had no or distribution of weight for age. The median age at the first measurable antibody in the previous serum specimen or a ;::4­ vaccine dose was 1 day in Oman and Thailand but 19 days in fold rise in antibody titer between the projected residual level of The Gambia, reflecting the fact that most children in The Gam­ maternally derived antibody and the infant's specimen, assuming bia are born at home. The median age at the 14-week visit was that maternal antibody decays with a half-life of 28 days. Geomet­ 112 days in The Gambia, 115 days in Oman, and 100 days in ric mean titers (GMTs) were calculated for all seropositive sub­ Thailand. The proportion of males was 50% in The Gambia, jects; a titer>1024 was taken to be equal to 2048. Virus isolationfrom challenge stool specimens. Virus isolation 53% in Oman, and 50% in Thailand. studies were done at Mahidol University, Bangkok; the Medical Serum specimens obtained at 24 weeks were available for Research Council Laboratories, Fajara, The Gambia; and Ruchill analysis for 1291 infants (77%) who had completed the immu­ Hospital, Glasgow, Scotland. All laboratories remained blinded as nization schedule (339 from The Gambia, 546 from Oman, and to the study subjects' identity and vaccination status. The three 406 from Thailand). Reasons for failure to complete the study laboratories followed a standard protocol for poliovirus isolation were moving away or inconvenience of travel to study clinic procedure on HEp-2 (Cincinnati) cells [3l}. The Study Group (141 infants in The Gambia, 4 in Oman, 138 in Thailand); Laboratory Coordinator visited each laboratory to monitor adher­ refusal ofparent to allow further blood collection (9 infants in ence to the protocol. Poliovirus isolates were typed using reference The Gambia, 37 in Oman, 3 in Thailand); withdrawal by study antisera prepared by the Rijksinstituut voor Volksgezondheid en physicians for illness (9 infants in Oman, 2 in Thailand); immu­ Mileiuhygiene, Bilthoven, Netherlands. Nonpolio enteroviruses nizations received outside study clinic (6 infants in Oman); were not typed. A sample of postchallenge stools from all sites reported as containing polioviruses was examined at Ruchill Hos­ death (2 infants in The Gambia, 4 in Oman); incomplete identi­ pital with Sabin strain-specific probes using a polymerase chain fication of forms or specimens (2 infants in The Gambia, 11 reaction (PCR) method [32]. in Oman); interval between doses >91 days (12 infants in The Statistical methods. X2 test, Fisher's exact test, Mantel-Haens­ Gambia, 2 in Oman, 1 in Thailand); interval between doses zel X2 test for linear trend, and Student's t test were used to detect <22 days (11 infants in Oman).

GAMBIA OMAN THAILAND 100%

a> > 80% :;:'w 0 c, 60% 0 L- a> 40% Figure 1. Seroprevalence to po­ en liovirustypes 1, 2, and 3 at 24 weeks of age (10 weeks after the last vac­ 20% cine dose), by vaccination status, in The Gambia, Oman, and Thailand. 0% 1 2 3 1 2 3 1 2 3 Poliovirus type

Vaccination status ~4 OPV .4 OPV + 3 IPV 03 IPV JID 1997; 175 (Suppl1) Infant Immunization with OPV and IPV S219

Table 3. Cumulative seroconversion at 14 and 24 weeks of age to poliovirus types 1,2, and 3 in infants immunized with OPV alone, OPV and IPV, or IPV alone, Oman.

Between birth and 14 weeks Between birth and 24 weeks

doses of 3 doses of OPV 4 doses of OPV Poliovirus 3 doses of OPV + 2 doses of IPV 2 doses of IPV 4 doses of OPV + 3 doses of IPV 3 doses of IPV

Type 1 122/150 (81) 112/138 (81) 96/136 (71)* 141/150 (94) 133/138 (96) 123/136 (90) Type 2 145/150 (97) 136/138 (99) 113/136 (83)t 148/150 (99) 137/138 (99) 131/136 (96) Type 3 109/150 (73) 126/138 (91)t 110/136 (81) 121/150 (81) 134/138 (97)~ 129/136 (95)~ Downloaded from https://academic.oup.com/jid/article/175/Supplement_1/S215/878782 by guest on 24 September 2021

* P < .05 compared with 3 doses of OPV, Fisher's exact test. t P < .01 compared with 3 doses of OPV, Fisher's exact test. ~ P < .01 compared with 4 doses of OPV, Fisher's exact test.

Seroprevalence. Seroprevalence findings at 24 weeks of and 92% for types 1, 2, and 3. Rates in the IPV group were age are shown in figure 1. In The Gambia, the OPV group significantly lower (P <.01) than rates in the referent OPV seroresponse rates at 24 weeks were 88% for type 1, 97% for group for all three types. type 2, and 72% for type 3. The combined-schedule group had Seroconversion. Seroconversion rates, by dose, could be seroresponse rates of97%, 100%, and 99%, and the IPV group calculated for Oman (table 3) and Thailand (table 4). had seroresponse rates of 81%, 82%, and 98% for types 1, 2, In Oman, between birth and 14 weeks of age, 81%, 97%, and and 3. The combined schedule performed significantly better 73% of the OPV group seroconverted to poliovirus types 1, 2, than the referent OPV group for type 1 (P < .01) and type 3 and 3, respectively (table 3). This compares with 81%,99%, and (P < .01). The IPV group had seroresponse rates significantly 91% seroconversion rates in the combined-schedule group and lower than the referent OPV group for type 2 (P < .01) but 71%,83%, and 81% in the IPV group. The seroconversion rate significantly higher for type 3 (P < .01). in the combined-schedule group was significantly higher than that In Oman, the OPV group seroresponse rates at 24 weeks in the OPV group for type 3 (P < .01). The seroconversion rates were 90% for type 1,98% for type 2, and 73% for type 3 (figure in the IPV group were significantly lower thanthose in the OPV 1). The combined-schedule group had seroresponse rates of group for type 1 (P < .05) and type 2 (P < .01). In Oman, by 95%, 99%, and 97%, and the IPV group had rates of 88%, 24 weeks of age, 94%, 99%, and 81% of children in the OPV 92%, and 91% for types 1,2, and 3, respectively. The serore­ group, 96%, 99%, and 97% in the combined-schedule group, and sponse rates were higher in the combined-schedule group than 90%, 96%, and 95% in the IPV group had seroconverted to in the referent OPV group for type 1 (P = .05) and type 3 (P poliovirus types 1, 2, and 3, respectively. In Oman, the serocon­ < .01). The IPV group had a significantly lower seroresponse version rates by 24 weeks did not differ between groups for types rate than the referent OPV group for type 2 (P < .05) but a 1 and 2, but the rates for type 3 seroconversion were significantly significantly higher response rate for type 3 (P < .01). higher (P < .01) in the IPV and combined-schedule groups com­ In Thailand, the OPV group seroresponse rates at 24 weeks pared with the OPV group. were 98% for type 1, 100% for type 2, and 100% for type 3 In Thailand, between birth and 14 weeks of age, 94%, 99%, (figure 1). The combined-schedule group had almost identical and 93% of children in the OPV group, 94%, 99%, and 95% in seroresponse rates of 99%, 100%, and 100% for types 1, 2, the combined-schedule group, and 40%, 48%, and 79% in the and 3. The IPV group had seroresponse rates of 66%, 63%, IPV group had seroconverted to poliovirus types 1, 2, and 3,

Table 4. Cumulative seroconversion at 14 and 24 weeks of age to poliovirus types 1,2, and 3 in infants immunized with OPV alone, OPV and IPV, or IPV alone, Thailand.

Between birth and 14 weeks Between birth and 24 weeks

3 doses of OPV 4 doses of OPV Poliovirus 3 doses of OPV + 2 doses of IPV 2 doses of IPV 4 doses of OPV + 3 doses of IPV 3 doses of IPV

Type 1 117/125 (94) 120/128 (94) 56/141 (40)* 123/125 (98) 127/128 (99) 94/141 (67)t Type 2 124/125 (99) 127/128 (99) 67/141 (48)* 125/125 (100) 127/128 (99) 92/141 (65)t Type 3 116/125 (93) 121/128 (95) 111/141 (79)* 125/125 (100) 127/128 (99) 132/141 (94)t

* P < .01 compared with 3 doses of OPV, Fisher's exact test. t P < .01 compared with 4 doses of OPV, Fisher's exact test. S220 WHO Collaborative Study Group JID 1997; 175 (Suppl 1) respectively (table 4). Seroconversion rates were strikinglylower Table 5. Geometric mean titer (95% confidence interval) of serum for the IPV group, and this difference was significant (P < .01) neutralizing antibody to poliovirus types 1, 2, and 3 at 24 weeks of age, in infants immunized with OPV alone, OPV and IPV, or IPV for all three types. In Thailand, between birth and 24 weeks of alone in The Gambia, Oman, and Thailand. age, the cumulativeseroconversion rates were ;:::98% for all three types in both the OPV and combined-schedule groups. Seroconv­ Site, 4 doses of OPV ersion rates for all three types were significantly lower (P < .01) poliovirus 4 doses of OPV + 3 doses of IPV 3 doses of IPV in the IPV group than in the OPV group, with 67% for type 1, The Gambia 65% for type 2, and 94% for type 3. Type 1 386 (297-503) 418 (327-533) 79 (56-113) Between countries, cumulative seroconversion rates by 24 n = 100 n = 115 n = 87 Downloaded from https://academic.oup.com/jid/article/175/Supplement_1/S215/878782 by guest on 24 September 2021 weeks differed for the OPV and IPV groups. The OPV group Type 2 597 (492-723) 744 (642-863) 144 (103-201) in Oman had a low rate for type 3 seroconversion and the IPV n = 111 n = 118 n = 88 group in Thailand had low rates of seroconversion for types 1 Type 3 133 (100-178) 341 (268-434) 241 (181-321) n = 82 n = 117 n = 105 and 2. In the OPV group in Oman, 94%, 99%, and 81% of Oman children seroconverted to poliovirus types 1, 2, and 3, respec­ Type 1 375 (305-460) 474 (381-589) 447 (362-550) tively, after four doses, compared with 98%, 100%, and 100% n = 171 n = 164 n = 161 in Thailand. In the IPV group in Oman, 90%, 96%, and 95% Type 2 718 (620-830) 982 (888-1085) 571 (467-697) of children seroconverted to poliovirus types 1, 2, and 3 after n = 187 n = 171 n = 169 Type 3 165 (132-206) 391 (325-471) 251 (197-319) three IPV doses, compared with 67%, 65%, and 94% in Thai­ n = 140 n = 166 n = 166 land. Thailand For the combined-schedule group, seroconversion rates at Type 1 681 (559-829) 840 (728-970) 49 (36-66) 14 weeks were higher in Thailand compared with Oman, but n = 126 n = 132 n = 95 by 24 weeks, children in both countries showed similar high Type 2 822 (721-938) 909 (798-1036) 68 (48-97) n = 128 n = 133 n = 92 rates of seroconversion. At 14 weeks (after a total of five Type 3 348 (282-429) 730 (635-839) 136 (103-179) polio vaccine doses), 81%, 99%, and 91% of children had n = 128 n = 133 n = 134 seroconverted to types 1, 2, and 3, respectively, in Oman and 94%, 99%, and 95%, respectively, in Thailand. At 24 weeks (after a total of seven polio vaccine doses), 96%, 99%, and ule group. A different pattern emerged for IPV, with signifi­ 97% had seroconverted to types 1, 2, and 3 in Oman and 99% cantly higher GMTs for types 1 and 2 in Oman. In the IPV to all three types in Thailand. group, the GMTs for types 1 and 2 in Oman were 4 or 5 times Geometric mean titers. GMTs at 24 weeks of age with higher than those in The Gambia and 8 or 9 times higher than their 95% confidence intervals are presented in table 5. those in Thailand. In The Gambia, GMTs for poliovirus type 1 were signifi­ Influence ofmaternal antibody on seroconversion in Oman cantly higher in the OPV group (386) and the combined-sched­ and Thailand. Based on cord blood specimens, maternal anti­ ule group (418) compared with the IPV group (79). For type body levels for Oman and Thailand were classified as high 2, GMTs were significantly higher for the OPV group (597) (titer ;:::64) or low (titer <64). There was a greater proportion and combined-schedule group (744) compared with the IPV of children with high levels of maternal antibody for,all three group (144). For type 3, the GMT was significantly lower in types in Oman than in Thailand. Among the 619 cord blood the OPV group (133) compared with the IPV group (241) and specimens from Oman, 62%, 57%, and 32% had high antibody the combined-schedule group (341). titers against poliovirus types 1, 2, and 3, respectively. In con­ In Oman, GMTs of neutralizing antibody to type 1 at 24 trast, among the 411 cord blood specimens from Thailand, weeks were not significantly different among the study groups. 41%, 38%, and 19% had high antibody titers against types 1, For type 2, GMTs in the OPV group (718) and the IPV group 2, and 3. (571) had overlapping 95% confidence intervals, but the GMT Seroconversion data were stratified according to level of was significantly higher in the combined-schedule group (982). maternal antibody. In the OPV and combined-schedule groups For type 3, the GMT was significantly lower in the OPV group in both countries, a high level of maternal antibody had little (165) and the IPV group (251) compared with the combined­ effect on the seroconversion rate between birth and 24 weeks schedule group (391). of age (data not shown). In Oman, the IPV group seroconver­ In Thailand, GMTs for all three serotypes were significantly sion rate was significantly lower for type 1 (P < .01) and type lower in the IPV group compared with the OPV and combined­ 3 (P < .05) in children with high levels of maternal antibody schedule groups. (figure 2). In Thailand, the IPV group seroconversion rates Across countries, in both the OPV group and the combined­ were 80%, 80%, and 96% against poliovirus types 1, 2, and 3 schedule group, the GMTs for types 1 and 3 were 2-fold higher in children with low levels ofmaternal antibody but only 45% in Thailand compared with The Gambia and Oman. Type 2 (P < .01), 39% (P < .01), and 82% (P < .05) in children GMTs were similar in the OPV group and the combined-sched- with high levels of maternal antibody. lID 1997; 175 (Suppl 1) Infant Immunization with OPV and IPV S221

OMAN THAILAND culture methods were selected for the PCR test. These included all type 2 and type 3 isolates, as well as 22 type 1 isolates 100% from all three countries. The PCR test detected Sabin strain polioviruses in 29 (97%) of the 30 stools and confirmed the c poliovirus type reported. In 3 stools, two poliovirus types were 0 80% - -- '00 identified with the Sabin-specific PCR, whereas neutralization "- (1) tests detected only one type. Nonpolio enteroviruses were iso­ c> 60% ..... lated from 14% of the postchallenge stools in each country, 0 0 with no difference between study groups in any of the study Downloaded from https://academic.oup.com/jid/article/175/Supplement_1/S215/878782 by guest on 24 September 2021 0 "- sites. (1) 40% - - en In The Gambia, 4% ofthe OPV group, 9% ofthe combined­ schedule group, and 16% ofthe IPV group shed type 1 poliovi­ 20% _. _ .. rus after challenge; the difference between the IPV group and the OPV group was significant (P < .01) (figure 6). In Oman, the challenge was administered after mass campaigns with OPV 0% 1* 2 3** 1* 2* 3** had been initiated in districts surrounding the study site, and the overall rate of shedding type 1 virus was 12%, with no Poliovirus type significant difference between vaccine groups. In Thailand, 14% of the OPV and combined-schedule groups shed type 1 Maternal antibody titer poliovirus after challenge, compared with 57% of the IPV group; the difference between the IPV group and the OPV .Iow «64) ~ high (>=64)

Figure 2. Seroconversion by 24 weeks of age in children who re­ ceived IPV at 6, 10, and 14 weeks of age, by level of maternal antibody. * P < .01; **P < .5.

Influence of OPV mass campaigns on seroconversion in Oman. Seroconversion data from Oman were stratified to ex­ amine the effect of OPV mass campaigns targeted at children 0-5 years of age in 38 districts around the study district. In response to an outbreak of wild type 3 poliovirus, mass cam­ paigns were conducted from April to December 1991, but the study district was not included (figure 3). At 6 weeks of age, the children in the IPV group had received only placebo, yet 32%, 46%, and 8% ofthose whose blood was drawn after the start of the mass campaigns had seroconverted for types 1, 2, and 3 (figure 4). This compared with 10%, 18%, and 8% of children whose blood was drawn before the date when OPV mass campaigns began. These differences were significant (P Indian ocean < .05) for types 1 and 2. Between birth and 10 weeks of age, these differences were even more striking, with significantly higher (P < .05) levels of seroconversion for types 1 and 2 in IPV vaccinees with secondary exposure to the mass campaign Studysitewith noOPV compared with those not exposed (64% vs. 27% seroconversion • masscampaigns for type 1, 82% vs. 41% for type 2) (figure 5). f':"iI Areaswhere.OPV ~ masscampaigns Challenge. Challenge virus isolation data were analyzed wereconducted for 1293 (77%) infants (334 in The Gambia, 561 in Oman, April-December 1991 and 398 in Thailand). Altogether there were 221 polioviruses Figure 3. Map of Oman showing study district (where there were isolated from the postchallenge stool specimens: 213 were type no OPV mass campaigns) and other districts where OPV mass cam­ 1, 6 were type 2, and 2 were type 3 polioviruses (table 6). paigns were conducted for children 0-5 years of age, April-Decem­ Thirty of the 221 stools with poliovirus detected by tissue ber 1991. S222 WHO Collaborative Study Group JID 1997; 175 (Suppl 1)

100% TYPE 1 TYPE 2 TYPE 3 c .iiio 80% s... Q) i; 60% o o Figure 4. Seroconversion to po­ e 40% liovirus types 1, 2, and 3 between enQ) birth and 6 weeks of age among chil­ Downloaded from https://academic.oup.com/jid/article/175/Supplement_1/S215/878782 by guest on 24 September 2021 dren not exposed compared with 20% children secondarily exposed to OPV mass campaigns, by vaccine 0%'---..Ll.UJ; group, Oman. * P < .05. 1 OPV Placebo" 1 OPV Placebo· 1 OPV Placebo 1 OPV + 0 IPV 1 OPV + 0 IPV 1 OPV + 0 IPV

OPV mass campaign status 1m No exposure • Secondary exposure

group was significant (P < .01). Among children in the IPV schedule with simultaneous delivery of OPV and IPV led to group who were seropositive to type 1 on the day of the chal­ excellent serologic response in all three countries and levels lenge, the proportion shedding type 1 poliovirus after challenge of mucosal protection comparable to those following adminis­ was 14% (12/86) in The Gambia, 10% (15/157) in Oman, and tration of OPV alone. The combined-schedule response was 54% (51/94) in Thailand (P < .01). not affected by socioeconomic status or level ofmatemal anti­ body. The variable response to OPV was consistent with previ­ ous reports, with lower response rates to type 1 and type 3 in Discussion children oflower socioeconomic status. As reported previously, The multicenter nature of this study allowed definition of high levels of maternal antibody did not affect the response to responses to three different polio immunization schedules OPV at 24 weeks. The secondary spread of OPV in the mass among children in three developing countries. The combined campaign setting was strongly suggested in Oman. The IPV

TYPE 1 TYPE 2 TYPE 3 100%

c o 80% '(j) s... Q) i; 60% o o Figure 5. Seroconversion to po­ e 40% liovirus types 1, 2, and 3 between Q) birth and 10 weeks of age among en children not exposed compared with 20% children secondarily exposed to OPV mass campaigns, by vaccine 0%) '------'-'-'- group, Oman. * P < .05. 20PV 1 IPV· 20PV 1 IPV· 20PV 11PV 2 OPV + 1 IPV 2 OPV + 1 IPV 2 OPV + 1 IPV

OPV mass campaign status 1m No exposure • Secondary exposure JID 1997; 175 (Suppl 1) Infant Immunization with OPV and IPV S223

Table 6. Results of virus isolation studies on postchallenge stools health care contacts.No interference was observed when children in The Gambia, Oman, and Thailand. received a dose of live vaccine and a dose of killed vaccine at the same visit. The serologic response appeared as good as that Gambia Oman Thailand Virus isolated No. (%) No. (%) No. (%) seen with two appropriately spaced doses of either vaccine. The type 1 mucosal protection afforded by the combined schedule Poliovirus type 1 31 (9) 65 (12) 117 (29) was as robust as that achieved with OPV alone. Poliovirus type 2 2 (1) 3 (1) 1 Although not addressed in this study, an added benefit of Poliovirus type 3 o 1 1 the combined schedule might be a lower risk ofOPV-associated Nonpolio enterovirus 46 (14) 80 (14) 56 (14) Toxic specimen 14 (4) o o poliomyelitis in recipients. First, vaccine doses are adminis­ Downloaded from https://academic.oup.com/jid/article/175/Supplement_1/S215/878782 by guest on 24 September 2021 No virus isolated 241 (72) 413* (73) 223 (56) tered at very young ages, when many infants are still protected Total 334 562 398 by maternal antibody. Second, high levels of serum antibody induced by IPV might prevent some OPV-associated disease * Includes 1 stool collected <2 days after challenge, which was not consid­ ered for final analysis. in recipients. Response to OPV. The serologic response to OPV in The Gambia and Oman was consistent with that reported in previous immunization schedule used in this study had shorter intervals studies in these countries [6, 29, 33]. After four OPV doses, between doses and an earlier starting age than most previous the percentage ofchildren with antibody titers :;::: 8 was low for studies. The serologic response to IPV in this abbreviated deliv­ type 1 (88% in The Gambia, 90% in Oman) and type 3 (72% ery schedule was lower than that reported in many previous in The Gambia, 73% in Oman). These results indicate that studies, and the response to IPV appeared to be affected by four doses of OPV cannot be relied on to provide total polio the level of maternal antibody. immunity. Even after four OPV doses, serum antibody gaps Response to the combined schedule. This study provides remain for type 1 and especially for type 3. unique data on the response of infants to a combined polio There was a much better serologic response to OPV in Thai­ immunization schedule with administration of OPV at birth land, with seropositivity rates of 98%, 100%, and 100% for and both OPV and IPVat 6, 10, and 14 weeks ofage. There was types 1, 2, and 3, respectively, after four doses. The response excellent serum antibody response to the combined schedule in in Thailand resembles the response in industrialized countries all three countries, indicating that the combined schedule may [5] and in higher-socioeconomic-status communities in devel­ provide a means for uniformly inducing high levels of serum oping countries [29, 34-36]. In the present study, children antibody with the primary immunization series. The combined enrolled in Thailand were from urban middle-socioeconomic­ schedule using DTP-IPV posed no operational or logistic prob­ status families, while those in The Gambia and Oman were lems in terms ofvaccine delivery: Neither extra visits nor extra from rural lower-socioeconomic-status families. Socioeco­ injections were required. There was no evidence of problems nomic status probably reflects critical levels of sanitation, hy­ with vaccine safety among the > 500 children in this study giene, and nutrition and is the best explanation for the better who received the combined schedule. Other implications for response to OPV in Thailand. Alternative (though less likely) global polio eradication (e.g., vaccine cost and availability) were not addressed in this study. More complex combined schedules have been adopted by some countries or areas, in­ 57% cluding Denmark, Iceland, Israel, Hungary, Lithuania, and the Vaccine group Palestine Self-Rule Areas of Gaza and the West Bank. Other ~40PV recent studies in Ivory Coast and The Gambia have examined the effect of giving a single dose of IPV at the same visit as .4 OPV + 31PV measles vaccine [28, 33]. While a schedule with three OPV D31PV doses followed by one IPV dose leads to higher levels of seroresponse compared with four OPV doses, the benefit is delayed until 9 months ofage. In the combined schedule tested in the current study, all polio vaccine doses were delivered by 14 weeks (3.5 months) of age. The combined schedule provided seven polio vaccine doses, compared with four doses in the OPV schedule and three doses in the IPV schedule. The response of children at 14 weeks in the Gambia* Oman Thailand* combined-schedule group after five polio vaccine doses (three Figure 6. Percentage of children shedding poliovirus type 1 in stool OPV and two IPV) was superior to the response at 24 weeks to 1 week after challenge with type 1 OPV, by vaccine group (numerals four doses of OPV or three doses of IPV. Thus, the combined refer to nos. of doses), in The Gambia, Oman, and Thailand. * Sig­ schedule led to higher antibody titers at an earlierage with fewer nificant difference (P < .01) between OPV and IPV groups. S224 WHO Collaborative Study Group JID 1997;175 (Suppll)

explanations for the results in Thailand are that they may reflect Intensive efforts were made to assure that vaccine lot, vaccine racial and genetic differences, that Thai children did not receive stability, and data errors were not the cause for the observed hepatitis B vaccine during the study (children in The Gambia site-to-site variation. Once IPV is combined with DTP, it is and Oman did), or that Thai children received infant fluoride not possible to use in vitro potency tests. Although in vivo supplements (children in The Gambia and Oman did not). Re­ assays could be used, they are relatively imprecise, and small cent malaria studies have demonstrated genetic differences in differences in potency probably could not be detected. There­ HLA type and T cell responses between persons in The Gambia fore, the potency of IPV field specimens was not examined in and Thailand [37, 38]. However, the low response to IPV in this study. However, the maintenance of vaccine potency in

Thailand does not suggest that humoral immunity to these epi­ field samples of OPV from all three study sites is reassuring. Downloaded from https://academic.oup.com/jid/article/175/Supplement_1/S215/878782 by guest on 24 September 2021 topes was unusually high. Previous studies of simultaneous Another concern was that two lots of IPV were used in this delivery of hepatitis B vaccine and either OPV or IPV do not study: One lot was used in The Gambia and Oman, and another suggest that hepatitis B vaccine has any effect on seroresponse lot was used in Oman and Thailand. To assess the potential to OPV or IPV [39-41]. Although fluoride has been found to impact of using different lots, Omani study participants were have an adjuvant effect on mucosal immunity in rats, no data stratified into groups depending on the lot ofIPV they received. have been published suggesting a similar effect in humans [42]. No significant differences were detected in seroprevalence or The wild type 3 outbreak during the study in Oman provided in seroconversion rates between these groups (data not shown). an unexpected opportunity to examine the secondary spread of These findings are consistent with the conclusions of the po­ OPV delivered in mass campaigns. Mass campaigns delivered tency tests on these two lots ofIPV. Within the detection limits two doses of OPV to nearly 100% of children 0-5 years of ofpresently available tests for IPV potency, there appeared to age in 38 districts adjacent to the study district but not in the be no difference between the two lots; however, present meth­ study district itself. In the social and cultural setting of Oman, ods are not well standardized [26], and a WHO collaborative this study demonstrates that OPV from the mass campaigns study to better correlate the findings of in vivo and in vitro penetrated rapidly into the study district. This provides further potency tests for IPV is in progress [49]. support for the mass campaign strategy advocated by Sabin et In both The Gambia and Oman, the response of the study al. [43] and de Quadros et al. [4]. children to IPV may have been enhanced by secondary expo­ Response to IPV. The overall serologic response to IPV sure to OPV-related polioviruses shed by other children in the was 66%-88% for type 1, 63%-92% for type 2, and 91%­ family and the community. In rural Gambia and Oman, families 98% for type 3. These findings suggest that IPV as a single are large and intense social contact is common in the village, vaccine given at 6, 10, and 14weeks ofage provides inadequate at markets, and at health centers. In contrast, middle-socioeco­ serologic protection against poliovirus, especially for type 1. nomic-status infants in Bangkok are likely to be raised in nu­ These findings are consistent with several previous studies that clear families, most often as only children. Thai infants are showed that delivery of IPV at intervals shorter than 8 weeks cared for at home by their mothers or grandmothers, and day led to lower levels ofseroresponse and that early age at delivery care centers are not available. In Thailand, there was a very of first dose and high levels of maternal antibody were addi­ low response to IPV for types 1 and 2. Because most of the tional factors leading to lower responses [44, 45]. Thai infants were single children and their lifestyle would not The response to IPV in the study was highly sensitive to have allowed many opportunities for contact with other OPV­ maternal antibody level. There were significantly lower rates of immunized children, the response to IPV shown in Thailand seroconversion for all three poliovirus types in Thailand and for may possibly be closer to the true response to this vaccine poliovirus types I and 3 in Oman between birth and 24 weeks when given with short intervals and at early ages. The effect in children whose baseline maternal antibody titers were ;;;:0;64. of schedule and age on response to IPV is demonstrated by Previousstudieshave shown that IPV appearsmore sensitivethan the observation in a previous study in Thailand at the same OPV to the effect ofhigh maternal antibodytiters on reducing the clinic: 100% of children were seropositive for poliovirus types neutralizingantibodyresponseto vaccinegiven in the firstmonths 1, 2, and 3 after three DTP-IPV doses given at 2-month inter­ oflife [44-48]. Krishnan et al. [44] have shown that the negative vals starting at 8 weeks of age [50]. effect of maternal antibody on response to IPV given at early Mucosal immunity. The 14% nonpolio enterovirus isola­ ages is enhanced if there are shorter intervals between doses. In tion rate in all three virology laboratories provides assurance that study, the presence ofany maternal antibody before the first about the sensitivity of virus isolation; nonpolio enterovirus IPV dose led to seroconversion rates of 94%, 46%, and 56% isolation rates of ;;;:0; 10% are considered sufficiently sensitive againstpoliovirustypes 1,2, and 3, respectively,when three IPV for the HEp-2 (Cincinnati) cell line used in this study [51]. In doses were given at 4-week intervalscompared with 100%,97%, all three countries, both OPV and the combination schedule and 88%, respectively, when the three doses were given at 8­ provided relatively good mucosal protection, as demonstrated week intervals [44]. by the low rate of challenge virus excretion. In addition, the present study found site-to-site variation in Low rates of shedding of type 1 poliovirus after challenge the response to IPV, which has not been reported previously. among IPV vaccinees in The Gambia (16%) and Oman (10%) JID 1997; 175 (Suppl 1) Infant Immunization with OPV and IPV S225

may be due to previous exposure of these children to OPV­ where paralytic poliomyelitis cases continue to occur. Such a related polioviruses shed by siblings or other toddlers. The low schedule offers countries that are far along in polio eradication rates oftype I shedding are consistent with the type 1 serologic an opportunity to close remaining gaps in immunity, an im­ response among IPV vaccinees of 81% in The Gambia and portant consideration when wild poliovirus importation will 88% in Oman. In Thailand, the response was very different, remain a threat until global eradication is achieved. with 57% of IPV vaccinees shedding type 1 poliovirus after challenge, which is compatible with the poor serologic response Study Group Members to type 1 (66%) in Thai infants immunized with IPV. In Thai­

land, a high proportion of seropositive IPV vaccinees shed the The Gambia: B. M. Greenwood (Principal Investigator), A J. Downloaded from https://academic.oup.com/jid/article/175/Supplement_1/S215/878782 by guest on 24 September 2021 type 1 challenge virus, whereas this was not true in The Gambia Hall, M. G. Rowe, and H. C. Whittle (Co-Principal Investigators), or Oman. This finding suggests that IPV alone may not interrupt Medical Research Council Laboratories, Fajara (B.M.G. and transmission of polioviruses, even when individual serologic A.J.H. are currently with London School of Hygiene and Tropical protection has been achieved. Medicine; M.G.R. is with University of Wisconsin, Madison); M. George, Ministry of Health, Banjul. Oman: A J. M. Suleiman (Principal Investigator), A A K. Al-Ghassani, M. Elbualy, and Conclusions P. G. Malankar (Co-Principal Investigators), Department of Family and Community Health Programmes, Ministry of Health, Muscat. The variable response to four doses ofOPV observed in this Scotland: G. B. Clements, E. T. Hay, C. Nairn, and M. H. Riding, study is consistent with previous reports of low response rates Regional Virus Laboratory, Ruchill Hospital, Glasgow. Switzer­ to type 1 and especially type 3 in some developing countries. land: E. Bell (Project Laboratory Coordinator), B. P. Hull (Acting This finding highlights the need for supplemental OPV delivery Project Laboratory Coordinator), S. E. Robertson (Project Direc­ strategies to achieve the goal of polio eradication. One such tor), P. F. Wright (Project Co-Director), H. Zoffmann, Global supplemental strategy is national immunization days. By the Programme for Vaccines and Immunization, WHO, Geneva (E.B. end of 1994, national immunization days had been conducted died in November 1993; H.Z. died in September 1994; P.F.W. is in 36 countries or areas, with an additional 25 countries plan­ currently with Vanderbilt University, Nashville). Thailand: C. Aree, R. Samakoses, S. Simasathien (Co-Principal Investigator), ning their first nationwide campaigns in 1995 [52]. These mass Department of Pediatrics, Phramongkutkao Hospital, Bangkok; S. campaigns are usually targeted at all children < 5 years of age Migasena (Principal Investigator), P. Pitisuttitham, B. Ponrat, Vac­ and provide two doses of OPV 1 month apart. Data from the cine Trial Center, Faculty of Tropical Medicine, Mahidol Univer­ present study in Oman indicate secondary spread ofOPV deliv­ sity, Bangkok; S. Tantivanich, S. Tharavanij, Department of Mi­ ered in mass campaigns to young children living in a geographi­ crobiology and Immunology, Mahidol University, Bangkok; cally large "pocket" (a district) that did not participate in United States: S. L. Cochi, R. W. Linkins (Project Coordinator for the mass immunization. Another recent study in Cuba found Data Analysis), P. A Patriarca, R. W. Sutter, National Immuniza­ seroconversion in nonvaccinated infants, probably from their tion Program, CDC, Atlanta (P.AP. is currently with the US Food exposure to vaccine-related polioviruses as a result of OPV and Drug Administration, Bethesda, Maryland); L. Anderson, K. mass campaigns [7]. Maher, M. Pallansch, S. Penaranda, Division ofViral and Rickett­ Data from the current study show that a schedule with three sial Disease, CDC, Atlanta. doses ofIPV at 6, 10, and 14 weeks ofage provides inadequate protection against poliovirus. This is probably due to an early Acknowledgments starting age, with interference from maternal antibody, and to short intervals between doses. Other data from this study sug­ Vaccines were donated by Pasteur-Merieux Serums et Vaccins, gest that IPV alone may not interrupt transmission of poliovi­ Lyon, France, and SmithKline Beecham, Rixensart, Belgium. The investigators thank the members of the former Research and Devel­ ruses, as well as not provide uniform individual protection even opment Group of the WHO Expanded Programme on Immuniza­ in fully vaccinated persons. tion (W. Foege, Chairperson) for guidance and support; J. Milstien, In contrast, the combined schedule with simultaneous deliv­ WHO, for arranging potency testing of the vaccines; and D. Wood, ery of OPV and IPV at 6, 10, and 14 weeks of age led to National Institute for Biological Standards and Control, UK, for excellent serologic response in three different parts of the reviewing potency test results. Special appreciation is extended to world. This combined schedule provided levels of mucosal the many children and their families who participated in this study. protection comparable to those found following administration of OPV alone. This schedule provides a safe and logistically feasible strategy for achieving high levels of serologic and References mucosal protection in infants at an early age. The response 1. Wright PF, Kim-Farley RJ, de Quadros CA, et al. Strategies for the global to the combined schedule does not appear to be affected by eradication ofpoliomyelitis by the year 2000. N Engl J Med 1991;325: 1774-9. socioeconomic status or level ofmaternal antibody. This sched­ 2. Hull HF, Ward NA, Hull BP, Milstien JB, de Quadros C. Paralytic polio­ ule could be applied routinely or selectively in high-risk areas, myelitis: seasoned strategies, disappearing disease. Lancet 1994; 343: including urban slums, areas with poor sanitation, and areas 1331-7. S226 WHO Collaborative Study Group JID 1997; 175 (Suppl 1)

3. Patriarca PA, Foege WH, Swartz TA. Progress in polio eradication. Lancet 25. Expanded Programme on Immunization, Programme for Acute Respiratory 1993;342:1461-4. Infections, Global Blood Safety Initiative, UNICEF, Copenhagen. Prod­ 4. de Quadros CA, Andrus JK, Olive lM, Macedo CG, Henderson DA. Polio uct information sheets 1993/1994. l Othed. Geneva: World Health Orga­ eradication from the Western Hemisphere. Annu Rev Public Health nization, 1993; document no. WHOIUNICEFIEPI.TS/93.1. 1992; 13:239-52. 26. Minor P. Summary report of a meeting on the estimation of the potency 5. Patriarca PA, Wright PF, John 'I'J. Factors affecting the immunogenicity of inactivated poliovaccine. Biologicals 1990; 18:243-4. of oral poliovirus vaccine in developing countries: review. Rev Infect 27. Expanded Programme on Immunization. Polio: rectal tubes for specimen Dis 1991; 13:926-39. collection. Technet News: Logistics for Health Newsletter 1992;92.2: 6. World Health Organization Collaborative Study Group on Oral Poliovirus 9. Vaccine. Factors affecting the immunogenicity of oral poliovirus vac­ 28. Moriniere BJ, van Loon FPL, Rhodes PH, et al. Immunogenicity of a cine: a prospective evaluation in Brazil and The Gambia. J Infect Dis supplemental dose of oral versus inactivated poliovirus vaccine. Lancet Downloaded from https://academic.oup.com/jid/article/175/Supplement_1/S215/878782 by guest on 24 September 2021 1995; 171:1097-106. 1993; 341:1545-50. 7. Mas Lago P, Bravo JR, Andrus JK, et al. Lessons from Cuba: mass 29. Sutter RW, Patriarca PA, Suleiman AlM, et al. Paralytic poliomyelitis campaign administration of trivalent oral poliovaccine and seropreva­ in Oman: association between regional differences in attack rate and lence of poliovirus neutralizing antibodies. Bull World Health Organ variations in antibody responses to oral poliovirus vaccine. Int J Epide­ 1994; 72:221-5. miol 1993;22:936-44. 8. Patriarca PA, Laender F, Palmeira G, et al. Randomized trial of alternative 30. Wood DJ, Heath AB. The second international standard for anti-poliovirus formulations of oral poliovaccine in Brazil. Lancet 1988; 1:429-33. sera types 1,2 and 3. Biologicals 1992;20:203-11. 9. Expanded Programme on Immunization. Poliomyelitis outbreak, Bulgaria. 31. Expanded Programme on Immunization, Division of Communicable Dis­ Wkly Epidemiol Rec 1992;67:336-7. eases. Manual for the virological investigation of poliomyelitis. Geneva: 10. Deming M, Jaiteh KO, Otten MW, et al. Epidemic poliomyelitis in The World Health Organization, 1990; document no. WHOIEPI/CDSIPO­ Gambia following the control of poliomyelitis as an endemic disease. LIO/90.1. II. Clinical efficacy of trivalent oral polio vaccine. Am J Epidemiol 32. Yang CF, De L, Holloway BP, Pallansch MA, Kew OM. Detection and 1992; 135:393-408. identification of vaccine-related polioviruses by the polymerase chain 11. Expanded Programme on Immunization. Mass vaccination campaigns to reaction. Virus Res 1991;20:159-79. control the 1991-1992 poliomyelitis outbreak, Jordan. Wkly Epidemiol 33. Hanlon P, Hanlon L, Marsh V, et al. Serological comparisons of ap­ Rec 1993;68:222-3. proaches to polio vaccination in The Gambia [letter]. Lancet 1987; 1: 12. Slater PE, Orenstein WA, Morag A, et al. Poliomyelitis outbreak in Israel 800-1. in 1988: a report with two commentaries. Lancet 1990;335:1192-8. 34. BernkopfH, Medalie J, Yekutiel M. Antibodies to poliomyelitis virus and 13. Expanded Programme on Immunization. Poliomyelitis outbreak, 1992, socioeconomic factors influencing their frequency in children in Israel. Malaysia. Wkly Epidemiol Rec 1993;68:297-300. Am J Trop Med Hyg 1957;6:697-703. 14. Iman H, Ismail HJM, Lal M. Poliomyelitis in Malaysia: two confirmed 35. Pal SR, Banerjee G, Aikat BK. Serological investigation on endemicity cases after 6 years without polio. Ann Trop Paediatr 1993; 13:339-43. of poliomyelitis in Calcutta and in neighbouring rural area. Indian J 15. van Niekerk ABW, Vries JB, Baard J, Schoub BD, Chezzi C, Blackburn Med Res 1966; 54:507 -11. NK. Outbreak of paralytic poliomyelitis in Namibia. Lancet 1994; 344: 36. Swartz TA, Skalska P, Gerichter CG, Cockburn we. Routine administra­ 661-4. tion oforal polio vaccine in a subtropical area: factors possibly affecting 16. Sutter RW, Patriarca PA, Brogan S, et al. Outbreak of paralytic poliomyeli­ sero-conversion rates. J Hyg Camb 1972;70:719-26. tis in Oman: evidence for widespread transmission among fully vacci­ 37. Zevering Y, Houghten RA, Frazer IH, Good MF. Major population differ­ nated children. Lancet 1991;338:715-20. ences in T cell response to a malaria sporozoite vaccine candidate. Int 17. Robertson SE, Suleiman AJM, Mehta FR, Al-Dhahry SHS, EI-Bualy MS. Immunol 1990;2:945-55. Poliomyelitis in Oman: acute flaccid paralysis surveillance leading to 38. Zevering Y, Khamboonruang C, Rungruengthanakit K, et al. Life-spans early detection and rapid response to a type 3 outbreak. Bull World of human T-cell responses to determinants from the circumsporozoite Health Organ 1994;72:907-14. proteins of Plasmodium falciparum and Plasmodium vivax. Proc Natl 18. Ramia S, ArifM. Paralytic poliomyelitis outbreak in Gizan, Saudi Arabia. Acad Sci USA 1994;91:6118-22. J Trop Pediatr 1991;37:202-4. 39. Coursaget P, Relyveld E, Brizard A, et al. Simultaneous injection of hepati­ 19. Vidor E, Caudrelier P, Plotkin S. The place ofDPT/eIPV vaccine in routine tis B vaccine with BCG and killed poliovirus vaccine. Vaccine 1992; paediatric vaccination. Rev Med Virol 1994;4:261-77. 10:319-21. 20. Petersen I. Control of poliomyelitis with inactivated and oral vaccines. 40. Giammanco G, LiVolti S, Mauro L, et al. Immune response to simultane­ Copenhagen: WHO Regional Office for Europe, 1991; document no. ICPIEPI/026/9. ous administration of a recombinant DNA hepatitis B vaccine and multi­ 21. Goldblum N, Gerichter CB, Tulchinsky TH, Melnick JL. Poliomyelitis ple compulsory vaccines in infancy. Vaccine 1991;9:747-50. control in Israel, the West Bank and Gaza Strip: changing strategies 41. King GE, Hadler SC. Simultaneous administration of childhood vaccines: with the goal of eradication in an endemic area. Bull World Health an important public health policy that is safe and efficacious. Pediatr Organ 1994; 72:783 -96. Infect Dis J 1994; 13:394-407. 22. Expanded Programme on Immunization. Global Advisory Group (1988). 42. Butler JE, Satam M, Ekstrand J. Fluoride: an adjuvant for mucosal and Wkly Epidemiol Rec 1989;64:5-10. systemic immunity. Immunol Lett 1990; 26:217 -20. 23. Expanded Programme on Immunization. EPI Information System, March 43. Sabin AB, Ramos-Alvarez M, Alvarez-Amezquita J, et al. Live, orally 1995. Geneva: World Health Organization, 1995; document no. WHO/ given poliovirus vaccine: effects of rapid mass immunization on popula­ EPVCEIS/95.1. tion under conditions of massive enteric infection with other viruses. 24. WHO Collaborative Study Group on Oral and Inactivated Poliovirus Vac­ JAMA 1960; 173:1521-6. cines. Response to an infant immunization schedule combining oral and 44. Krishnan R, Jadhav M, John 'I'J. Efficacy of inactivated poliovirus vaccine inactivated poliovirus vaccines compared with either vaccine alone: in India. Bull World Health Organ 1983;61:689-92. results of a randomized trial in The Gambia, Oman, and Thailand: final 45. Simoes EAF, Padmini B, SteinhoffMC, Jadhav M, John TJ. Antibody report. Geneva: World Health Organization, 1995; document no. WHO/ response of infants to two doses of inactivated poliovirus vaccine VRD/GEN/95.1. of enhanced potency. Am J Dis Child 1985; 139:977-80. JID 1997;175 (Suppl I) Infant Immunization with OPY and IPY S227

46. Faden H, Modlin JF, Thoms ML, McBean AM, Ferdon MB, Ogra PL. 49. WHO Expert Committee on Biological Standardization. Forty-fourth re­ Comparative evaluation of immunization with live attenuated and port. WHO Tech Rep Ser 1994;848:11. enhanced-potency inactivated trivalent poliovirus vaccines in child­ 50. Simasathien S, Migasena S, Beuvery C, et al. Comparison of enhanced hood: systemic and local immune responses. J Infect Dis 1990; 162: potency inactivated poliovirus vaccine (EIPY) versus standard oral po­ 1291-7. liovirus vaccine (OPY) in Thai infants. Scand J Infect Dis 1994;26: 47. Kok PW, Leeuwenburg J, Tukei P, et al. Serologicaland virological assess­ 731-8. ment of oral and inactivated poliovirus vaccines in a rural population 51. Expanded Programme on Immunization. Laboratory support for activities in Kenya. Bull World Health Organ 1992;70:93-103. aimed at the global eradication of poliomyelitis by the year 2000. Ge­ 48. McBean AM, Thoms ML, Albrecht P, Cuthie JC, Bernier R, Field Staff neva: World Health Organization, 1995; document no. WHO/EPI/GENI and Coordinating Committee. Serologic response to oral polio vaccine 95.4. and enhanced-potencyinactivated polio vaccines. Am J Epidemiol1988; 52. Expanded Programme on Immunization. Progress toward poliomyelitis Downloaded from https://academic.oup.com/jid/article/175/Supplement_1/S215/878782 by guest on 24 September 2021 128:615-28. eradication, 1994. Wkly Epidemiol Rec 1995;70:97-101.