Validation of the combined toxin-binding inhibition test for determination of neutralizing antibodies against tetanus and diphtheria toxins in a vaccine field study in Viet Nam.

Validation of the combined toxin-binding inhibition test for determination of neutralizing antibodies against tetanus and diphtheria toxins in a vaccine field study in Viet Nam H.A. Hong,1 N.T. Ke,2 T.N. Nhon,3 N.D. Thinh,3 J.W. van der Gun,4 J.T. Hendriks,5 & J.G. Kreeftenberg6 Determination of seroconversion and measurement of protective antibody levels in children against vaccine components are essential for gauging and monitoring the efficacy of paediatric vaccination programmes. For this purpose, we assessed the combined toxin-binding inhibition (ToBI) test for determining neutral- izing antibodies to tetanus and diphtheria in a diphtheria-pertussis-tetanus (DPT) vaccine field trial in Viet Nam. A simple procedure involving collection of blood samples on filter-paper was found to be a suitable alternative to collection by venepuncture, despite a reduction in the sensitivity of the ToBI test as a result of the step necessary to elute the antibodies from the filter-paper. The results obtained demonstrate that the ToBI test can feasibly be carried out under field conditions. Preliminary results obtained with the ToBI test in DPT field trials indicate that a fourth dose of DPT vaccine one year after the third dose should be considered by developing countries. Introduction WHO promotes improvement of quality control and quality assurance in those countries that meet the criteria for strengthening local vaccine production.a Viet Nam (population, 71 million) is such a country and has been selected by the Children's Vaccine Initiative (WHO/CVI) as a priority setting for rapid implementation of vaccine self-sufficiency programmes.b Diphtheria-pertussis-tetanus (DPT) adsorbed ' Head, Immunology Laboratory, Quality Control Department, Na- tional Institute of Vaccines and Biological Substances (IVAC), Nha Trang, Viet Nam. 2 Director, IVAC, Nha Trang, Viet Nam. 3 Laboratory Technician, Immunology Laboratory, Quality Control Department, IVAC, Nha Trang, Viet Nam'. 4 Senior Technician, Laboratory for the Control of Biological Prod- ucts, National Institute of Public Health and Environmental Protec- tion (RIVM), Bilthoven, Netherlands. 5 Deputy Coordinator, Bureau for International Cooperation, RIVM, P.O. Box 1, 3720 BA Bilthoven, Netherlands. Correspond- ence should be sent to Dr Hendriks at this address. 6 Senior Advisor on Vaccine Quality, Bureau for International Co- operation, RIVM, Bilthoven, Netherlands. a Expanded Programme on Immunization: vaccine quality. Report by the Director-General. Eighty-ninth session of Executive Board. WHO unpublished document EB89/18, 1991. b Children's Vaccine Initiative: report of the third meeting of the Consultative Group, Kyoto, Japan, 7-9 November 1993. WHO unpublished document CVI/CG-3/93 (pages 17 and 28). Reprint No. 5700 vaccine is one of the most important used in the Expanded Programme on Immunization (EPI) in Viet Nam. Traditionally this vaccine was prepared using a static culture method, which had the disad- vantage that the production capacity was not suffi- cient to meet the demand. Since 1986, however, a modern production line for DPT vaccines (sup- ported by UNICEF), based upon fermenter technol- ogy, has been operational in the National Institute of Vaccines and Biological Substances (IVAC) in Nha Trang and Dalat. After it had been confirmed in comparative du- plicate quality control evaluations that various DPT vaccine lots, locally produced at IVAC, met all WHO requirements for safety and potency, they were used in a field trial. UNICEF DPT vaccine, obtained through UNICEF/WHO-EPI in Viet Nam, was included in the trial for reference purposes. The combined toxin-binding inhibition (ToBI) test was selected to determine antibodies against diphtheria and tetanus toxins, because it has been properly validated for correlation with the in-vivo toxin neutralization test, also in the low titre range (1). Also, the ToBI test appears to be suitable for field trials since it requires only small amounts of blood; it has an established reproducibility and sensi- tivity, and it requires no experimental animals, un- like the in-vivo toxin neutralization test. Finally, because the ToBI test is an enzyme-linked immunosorbent assay (ELISA) it is suitable for deal- ing with large numbers of samples. Bulletin of the World Health Organization, 1996, 74 (3): 275-282 © World Health Organization 1996 275 H.A. Hong et al. A simple method involving collection of blood on filter-papers was employed for the following rea- sons: samples can more easily be transported under tropical conditions; and it avoids the difficulties as- sociated with collecting blood from infants by venepuncture. The main objective of the study was to validate the ToBI test and the use of a simple filter-paper blood collection method in a developing country. A further aim was to illustrate the test's prac- tical application in the preliminary assessment of the immunogenicity of locally produced DPT vaccine. Materials and methods Blood collection using filter-paper Blood was collected following a method developed by the National Institute of Hygiene and Epidemiol- ogy (NIHE), Hanoi, Viet Nam, and which is rou- tinely used in measles surveillance studies. Briefly, blood samples were collected from a finger using a needle and absorbed fully onto two squares (1 cm x 1.5 cm) of filter-paper (Whatman type 3). This en- abled 50,tl of blood to be collected (approximately 25 >t of serum). The papers were dried at room tem- perature, placed in a small light-coloured plastic bag, and sealed before being transported to a cool, dry place until testing. Before being sealed in the bags, the papers were carefully inspected to ensure that they were completely dry. Since all blood collections took place in the hot, dry season, the papers had usually dried within an hour, although in some cases an incubator was used. All filter-paper samples were coded. Prior to testing, each filter-paper was soaked in 250tl of phosphate buffer solution for 1 hour, with occasional shaking, to obtain a 1:10 dilution of the serum sample. In the combined ToBI test, 1:10 diluted serum was added to the first column of the wells. Combined ToBI test for diphtheria and tetanus neutralizing antibodies The ToBI test was used as described previously by Hendriksen et al. (2). Briefly, 100 tl of twofold dilu- tions of both reference and test sera were made up in phosphate-buffered saline. Subsequently, 120 [tl of a mixture containing 0.02LF units/ml of tetanus toxin and 0.01 LF units/ml of diphtheria toxin were added. After overnight incubation, any non-neutralized toxin was detected by ELISA by transferring 100 pl of the serum-toxin mixtures to immunoassay plates coated with tetanus or diphtheria antibodies. After the plates had been incubated, bound toxin was de- tected by successive incubation with biotinylated horse-antitetanus or horse-antidiphtheria and steptavidin biotinylated horseradish peroxidase complex followed by addition of 3,3',5,5'-tetra- methylbenzidine substrate. The enzymatic reaction was stopped after 10 minutes by addition of 100 t1 of 2-mol/I H2SO4. The absorbance in all the wells was measured using an automatic plate reader. Estimated antibody titres were multiplied by the pre-dilution factor (1:10). The relative antibody activity of each serum sample was calculated by comparing the OD50 values (the absorbance (A) below the value representing 50% of the total absorbance, defined as the sum of the arithmetic mean absorbance of positive control wells and the arithmetic mean absorbance of nega- tive control wells) of the sera under test with the OD50 of the National Institute of Public Health and Environmental Protection (RIVM) horse reference antitoxin. The results reported here (with the excep- tion of those shown in Fig. 1) were calculated using this OD50 method. The results shown in Fig. 1 were obtained using Biotek software (Biotek, Winooski VT, USA) to estimate the antibody levels against a calibration curve employing a four-parameter logistic fit. In this regression technique the anti- body activity of the test serum is calculated and expressed in IU/ml against the RIVM horse refer- ence antitoxin. Negative results, i.e., those below the detection level of the test, were set at 0.001 IU/ml. The mini- mum antibody level that could be detected by the ToBI test in this study was, however, 0.06 IU/ml. Serological field studies ToBI validation. Three main groups of samples were used for validation of the ToBI test as outlined below. * To validate the tetanus ToBI test used in IVAC against the ToBI test at RIVM, blood samples col- lected by venepuncture from 20 human volunteers (IVAC staff) were divided into two aliquots and tested blindly for tetanus antibodies at two different locations: the first was sent to RIVM, Bilthoven, Netherlands; the other was tested at IVAC, Nha Trang, Viet Nam. * To validate the ToBI test and the filter-paper method two experiments were carried out: 1) 22 blood samples were collected by finger-prick and venepuncture and tested at IVAC; this was carried out prior to the start of the field studies and the WHO Bulletin OMS. Vol 74 1996276 Combined toxin-binding inhibition test for tetanus and diphtheria toxins Fig. 1. Regression plots of neutralizing antibody titres determined using the ToBI method for blood samples collected using the filter-paper method ver- sus those collected by venepuncture. a) Antibodies against tetanus. b) Antibodies against diphtheria. Fig- ures in parentheses indicate the number of samples that were negative. 10 -J iP 0.1 ca. @0.01 0.001 b) 0.01 0.1 Serum/titre (IU/ML) 1 10 20 0.01 0.1 1 Serum/titre (IU/ML) results were calculated using the OD50 method; and 2) 49 blood samples collected by finger-prick and venepuncture and tested at IVAC using a new titre estimation method (Biotek). Clinical vaccine trial Study design. A DPT vaccine field trial is being car- ried out by the Vietnamese Ministry of Health to evaluate various lots of DPT vaccines, both those produced nationally and those supplied by UNICEF. The trial involves three phases, as described below, but only the results of the first phase are reported here. * Phase I: 15-30 infants per group (i.e., per vaccine lot) are studied mainly for side-effects and serology for the diphtheria and tetanus components. * Phase II: after evaluation of phase I results, 15-100 infants per group are studied mainly for side-effects and serology (diphtheria and tetanus). * Phase III: after evaluation of phase II results, 1000 infants per group are evaluated to obtain prelimi- nary information on the efficacy of the diphtheria and tetanus components of the vaccines studied. The use of EPI programme cards (one per child) proved helpful in the assessment of the immunization status of the infants. A total of 90 healthy unvaccinated infants aged 3-9 months with no symptoms of intestinal disease entered phase I of the study, which was performed in two distinct areas/provinces: Tien Giang and Lam Dong in the south of Viet Nam. The infants in each province were divided randomly into three groups, with about 15 infants in each group, corresponding to the three vaccines used in the study. Since the study was longitudinal, individual children could be followed throughout its course, allowing sero- conversion to be monitored. Vaccines, immunization schedule, and administra- tion. Each child received three intramuscular in- jections (0.5ml each of vaccine) at 1-month intervals. Three lots of DPT vaccine (one procured through UNICEF (the "UNICEF" vaccine) and two locally produced), all of which met WHO require- ments for safety and potency, were included in the study. Only data on two vaccines are reported here. The locally produced vaccine (lot DPT-23K), data for which are reported here, contained diphtheria and tetanus components that had been mixed with a pertussis component procured through UNICEF. Both the UNICEF and lot DPT-23K vaccines had the following composition: 16 IOU of pertussis, 15 LF of diphtheria toxoid, 10 LF of tetanus toxoid, and 1.5 mg of aluminium phosphate. Samples of peripheral blood were collected prior to each of the three injections, as well as 1 month and 1 year after the third injection. Collection was carried out by finger-prick using filter-papers. Statistical tests Logarithmic-transformed antibody titres obtained by two different blood collection methods were com- pared using regression analysis. Results Standardization of the ToBI test Table 1 shows the tetanus antibody titres estimated WHO Bulletin OMS. Vol 74 1996 a) - Positive * Negative (13) Y= 0.971x-0.101 r= 0.988 ' " "" " " 1,1 |1 1121 AI 0 277 H.A. Hong et al. using the ToBI test in the RIVM and IVAC labora- tories. The results demonstrate a high correlation between the two laboratories; regression analysis produced a correlation coefficient (r) of 0.922; Y = 0.967X - 0.21. A total of 10 samples were classified as "negative" (i.e., below the detection level of 0.06IU per ml) in both laboratories; eight samples were classified as significantly >0.1 IU per ml in both locations; in two cases a titre of 0.06 IU/ml in IVAC was found to be <0.06 IU/ml in RIVM. A repeat test carried out at RIVM on both these samples yielded a titre of 0.06 IU per ml for one of them. It was con- cluded that the inter-laboratory variation fell within that in each individual laboratory. Blood collection by the filter-paper method Since finger-prick collection of blood on filter-papers was held to be convenient for use in large-scale field trials, we evaluated the effect on the ToBI test of collecting blood this way compared with collection by venepuncture using samples from human volun- teers. Two validation studies were carried out as dis- cussed below. Table 1: Tetanus antibody titres in human sera estimated using the ToBI test in the two study laboratories Antibody titre in lU/ml at:a Sample code IVAC RIVM 01 0.125 0.125 02 2.0 4.0 03 1.0 1.0 04 2.0 1.06 05 <0.06 <0.06 06 0.25 0.5 07 1.0 2.0 08 <0.06 <0.06 09 <0.06 <0.06 10 <0.06 <0.06 11 0.06 <0.06b 12 <0.06 <0.06 13 0.06 <0.06b 14 <0.06 <0.06 15 <0.06 <0.06 16 <0.06 <0.06 17 0.5 0.25 18 <0.06 <0.06 19 1.0 1.0 20 <0.06 <0.06 a Shown are the titres for neutralizing anti-tetanus antibodies on paired samples from 20 human volunteers from Viet Nam. IVAC = National Institute of Vaccines and Biological Substances, Nha Trang, Viet Nam. RIVM = National Institute of Public Health and Environmental Protection, Bilthoven, Netherlands; <0.06 = too low to detect. b Significant discrepancy between both laboratories (see text for details). The first study, carried out prior to the start of the field studies, used 22 blood samples collected from human volunteers and the results were calcu- lated using the OD50 method. Regression analysis of the results showed that the correlation coefficient was high for both antibodies: tetanus (r = 0.98); and diphtheria (r = 0.95) (data not shown). Under the conditions chosen, the method can be used reliably in field studies to estimate antibody levels greater than 0.06 IU per ml; this detection limit is a conse- quence of the tenfold reduction in sensitivity caused by the elution of the antibodies from the filter-paper (see Discussion). Since antibody levels 0.01 IU/ml are generally held to provide clinical immunity against diphtheria and tetanus, a positive ToBI test indicates that a child is probably protected. The second validation study involved blood samples from 49 human volunteers using a novel method for estimating titres (Biotek). The results (Fig. 1) corroborate those found in the first study discussed above, i.e., in the ToBI test, blood col- lected on filter-paper gives similar titre estimations to those obtained by venepuncture. A regression analysis showed that the correlation coefficients were even higher for the Biotek method than for the OD50 approach for both antitoxins: tetanus (r = 0.998, n = 49); diphtheria (r = 0.956, n = 49). Children with antibody titres >0.061U/ml after vaccination As part of an ongoing phase-I DPT vaccine field study, in which different vaccine lots were being evaluated, the seroconversion and subsequent devel- opment of antibody titres among a group of 30 chil- dren immunized with vaccine lot 23K were followed up. The results obtained were compared with data for another group of 30 children immunized with the UNICEF vaccine. Two groups of 30 children were immunized with either lot DPT-23K or UNICEF DPT vaccine and followed up for 1 year after the third injection. Of the 30 infants immunized with lot DPT-23K, seven (23%) had antibody titres >0.06 IU/ml against tetanus, whereas four (13%) exhibited titres >0.06IU/ml against diphtheria before the first dose was given (pre-phase I) (Table 2). In particular, we examined the development of antibody titres among children who were seronegative at the pre-phase I stage. Table 2 shows the number of children (expressed as a % of the number of seronegative children at the pre-phase I stage) with antibody titres >0.06IU/ml up to 1 year after completion of the immunization schedule. For diphtheria and tetanus antitoxins, the proportion of children with titres >0.06IU/ml one 278 WHO Bulletin OMS. Vol 74 1996 Combined toxin-binding inhibition test for tetanus and diphtheria toxins Table 2: Distribution of study children (seronegative before the first dose of vaccine) with neutralizing antibody titres against tetanus and diphtheria after immunization with DPT-lot 23K or a reference UNICEF vaccinea Tetanus Diphtheria pre-l pre-l1 pre-lIl 1 mo-lIl 1 yr-Ill pre-l pre-l1 pre-lIl 1 mo-lIl 1 yr-Ill DPT-lot 23K No. of children 30 30 No. >0.06IU/ml at pre-l 23 22 21 21 17 26 25 24 23 20 % with titre <0.061U/ml 76.7 86.7 % with titre >0.061U/ml 9.1 100 100 100 4 58.3 100 45 % with titre >0.5 lU/mI 0 76.2 95.2 17.4 0 12.5 47.8 15 GMT (IU/ml)b 0.001 0.909 3.036 0.221 0.001 0.026 0.552 0.011 UNICEF-DPT No. of children 30 30 No. s0.061U/ml at pre-l 25 23 23 23 17 29 27 27 27 20 % with titre s0.06 lU/mI 83.3 96.7 % with titre >0.061U/ml 56.5 100 100 100 11.1 81.8 96.3 70 % with titre >0.5 lU/ml 0 73.9 100 58.8 0 22.2 57.7 10 GMT (IU/ml)b 0.014 0.920 5.153 0.496 0.002 0.113 0.492 0.033 a DPT = diphtheria-pertussis-tetanus; pre-l, pre-l1, pre-IIl = pre-immunization serum, pre-second dose, and pre-third dose, resp; 1 mo-Ill and 1-yr-Ill = 1 month post-third dose and 1 year post-third dose, resp. b GMT = geometric mean titre. month after the third injection was 100%. One year after the third injection, a significant reduction in the number of children with antibody levels >0.06 IU/ml occurred for diphtheria (9/20 (45%)), while for teta- nus 100% (17/17) still had antibody levels above this value. Similar trends were observed for different do- mestically produced DPT vaccine (data not shown) as well as the UNICEF vaccine supplied through the EPI programme (Table 2). It should be borne in mind, however, that the proportion of children actu- ally protected against clinical disease by the vaccines studied will be greater than the 45% reported here since the cut-off applied (>0.06 IU/ml) is well above the 0.01 IU/ml level that is believed to provide clini- cal immunity against diphtheria. The data in Table 2 show that for both diph- theria and tetanus the titres decreased between 1 month and 1 year after the third dose. At least for diphtheria, this may lead to a subsequent substantial increase in the number of children with antibody titres -,-0.06IU/ml. Discussion Field studies in developing countries are hampered by the lack of reliable, reproducible screening meth- ods that can be easily carried out under tropical conditions. Our findings confirm the feasibility of using the ToBI test as a screening tool in field trials for evalu- ation of vaccines. The test was implemented at the IVAC quality control laboratory by one of us (H.A.H.) after a period of training at RIVM, where it had been developed and standardized. Collection of blood on filter-papers was found to be satisfactory in combination with the ToBI test. Other workers have also recently reported that filter-paper collection is a reliable alternative to venepuncture for determination by ELISA of anti- toxin titres to tetanus and diphtheria in blood sam- ples. The particular advantage of the ToBI test, however, over such direct ELISA assays is its good correlation with the results of the in-vivo mouse toxin neutralization test in the low titre range (3). In order to obtain the cooperation of mothers, all efforts should be undertaken to minimize the "trauma" associated with blood collection from neonates. The filter-paper collection method in com- bination with the ToBI test appears to be the optimal choice. The method proved to be very reliable and reproducible, since aliquots of the same human adult serum samples tested at different locations yielded consistently the same results (see Table 1). Finally, from a public health point of view, the sensitivity level of 0.06 IU/ml used in this study pro- vides a safety margin compared with other studies, which have assumed that a titre of 0.01 IU/ml would provide clinical immunity against diphtheria. A novel method of estimating antibody activity, based upon a four-parameter fit (Biotek) (Fig. 1) indicated that with this method a cut-off -0.06 IU/ml can be defined (Fig. 1). Further study is required to deter- mine whether this approach in conjunction with the filter-paper/ToBI procedure would lead to reliable WHO Bulletin OMS. Vol 74 1996 279 H.A. Hong et al. Table 3: Comparison of the geometric mean titres of diphtheria and tetanus antibodies before and after the third dose of DPT vaccine between children in the present study and studies carried out in the USA Study: Present Present (lot 23 K (UNICEF Anderson et al. Anderson et al. Edwards et al. Barkin et al. Sample taken:" vaccine) vaccine) (ref. 10) (ref. 11) (ref. 12) (ref. 13) Diphtheria Pre-ll 0.001 0.002 0.04 0.01 0.06 Pre-IIl 0.03 0.11 0.23 0.31 0.03 0.07 1 Mo-IlIl 0.55 0.49 0.75 1.45 0.25 0.26 lYr-Ill 0.01 0.03 0.17 Tetanus Pre-l1 0.001 0.01 0.05 0.04 0.31 Pre-IIl 0.91 0.92 1.65 0.42 0.13 0.15 iMo-Ill 3.04 5.15 13.18 2.40 1.50 0.51 lYr-Ill 0.22 0.50 0.23 a See footnote a, Table 2. titre estimations in the range 0.06-0.01 IU/ml. The need for serological monitoring of vaccinees has been emphasized by studies in Sweden, which have reported that the protection level against diphtheria fluctuates considerably in individuals (4,5). Our pre- liminary findings confirm those reported in studies carried out in Africa, Europe, and the USA that antibody levels against the diphtheria component of DPT vaccine tend to decrease rather quickly (4, 6- 10). This is also illustrated in Table 3, which com- pares our geometric mean titres at certain times after immunization with those reported by various studies from the USA (10-13). Confirmation of our findings by a larger survey would provide a scientific basis for the recommenda- tions made by the Vietnamese Ministry of Health that a follow-up (fourth) booster dose in DPT immunization programmes should be given. This is in line with the general recommendation for diphtheria immunization policy in developing countries, i.e., should the epidemiology of diphtheria pose a significant health problem in preschool or school-age children, use of supplementary doses of diphtheria toxoid following high coverage primary immunization of under-1-year-olds is strongly favoured.c It has been proposed that fully immunized chil- dren with antibody titres against diphtheria that gradually decrease with time (eventually falling be- low 0.01 IU/ml) still possess circulating memory cells and that such "primed" children probably have a basic immunity that protects them from clinical dis- ease when challenged in nature (8).C Nevertheless, a c Galazka AM. The immunological basis for immunization (module 2: diphtheria). Unpublished document WHO/EPI/Gen/93.12, 1993. policy advocating use of a booster dose after primary vaccination may be wise since the current diphtheria epidemic in the Commonwealth of Independent States (CIS) has revealed that (mild) cases of diph- theria occurred in persons with a full immuniza- tion history. This may be a result of the absence of natural stimuli caused by the reduction in the Corynebacterium diphtheria reservoir, leading to a shorter post-vaccination immunity or, alternatively, may have been related to the suboptimal immune response following use of low potency vaccines for infant primary series immunization.d Implementation of the ToBI test in a laboratory engaged in routine quality control of vaccines has the additional advantage that it could serve as an in- process control to monitor the potency of diphtheria and tetanus toxoids. Use of the ToBI test in the T- potency test has been proposed (14). The successful introduction of the ToBI test under field conditions allows objective evaluation and comparison of locally produced vaccines, which is critical for the release by national control authori- ties of domestically produced vaccines to replace imported vaccines in EPI programmes. The evidence provided in Table 2 shows that 1 month after the third dose, the locally produced DPT-lot 23K con- forms in general with WHO requirements for diph- theria and tetanus vaccines, i.e., that for 90% of the target population antitoxin levels against diphtheria and tetanus should be >0.01IU/ml after the com- pletion of primary immunization of previously unvaccinated individuals (15). dDiphtheria prevention and immunization programmes in newly independent states: report on a WHO Meeting, Berlin, 18-20 Janu- ary 1995. Unpublished WHO document EUR/ICP/CMDS 94 06/ MT06, 1995. 280 WHO Bulletin OMS. Vol 74 1996 Combined toxin-binding inhibition test for tetanus and diphtheria toxins Finally, the work described here illustrates the usefulness of incorporating into technology transfer projects validated quality control methods as well as serological methods for disease surveillance in the target population. Acknowledgements Dr H. Rumke and Dr C. Hendriksen are thanked for their advice and critical reading of the manuscript. Financial support was provided by the Netherlands Ministry of De- velopment Cooperation. Resume Validation de I'epreuve associee d'inhibition de la liaison a la toxine pour la determination des anticorps neutralisants diriges contre les toxines tetanique et diphterique au Viet Nam La d6termination de la seroconversion et la mesure des taux d'antigenes protecteurs diriges contre les constituants des vaccins chez l'enfant sont des outils indispensables pour 6valuer et surveiller l'efficacite des programmes de vaccination pedia- trique. A cet effet, nous avons 6value l'epreuve associee d'inhibition de la liaison a la toxine (ToBI) pour la d6termination des anticorps neutralisants dirig6s contre les toxines t6tanique et diphterique, lors d'un essai pratique de vaccin antidiphterique- anticoquelucheux-antitetanique (DTC) au Viet Nam. L'6preuve a 6te realisee sur des pr6levements de sang recueillis sur papier filtre, m6thode directe qui facilite le transport des 6chantillons en milieu tropical et 6vite les problemes que pose le pr6leve- ment de sang par ponction veineuse chez le nourrisson. Le test ToBI est correctement valide du point de vue de sa correlation avec l'6preuve in vivo de neutralisation de la toxine, 6galement dans la gamme des faibles titres, et n'exige que de petites quantites de sang; sa reproductibilite et sa sen- sibilit6 sont 6tablies, et aucun animal d'experience n'est necessaire. D'apres nos r6sultats, le ToBI a 6te mis en ceuvre et valide avec succes au laboratoire de contr6le de la qualit6 de l'institut national de produc- tion de vaccins au Viet Nam. Le pr6levement d'6chantillons de sang sur papier filtre est une alternative valable a la ponction veineuse, malgre une perte de sensibilite li6e a l'6tape d'elution de l'anticorps a partir du papier filtre. Plusieurs 6tudes de validation a petite 6chelle ont confirm6 qu'avec cette 6preuve le prelevement de sang sur papier filtre donne des estimations du titre analogues a celles obtenues par ponction veineuse chez le meme individu (r = 0,988 pour le tetanos et 0,956 pour la dipht6rie). L'epreuve ToBI utilis6e en association avec le pr6levement de sang sur papier filtre repr6sente donc une approache valable et r6alisable de la surveillance de la reponse s6rologique aux con- stituants t6tanique et diphterique du vaccin DTC lors d'essais pratiques en milieu tropical. Les resultats preliminaires d'une etude en double aveugle utilisant l'approche ToBI/papier filtre in- diquent qu'une quatribme dose de DTC inject6e un an apres la troisieme dose pourrait etre envisagee dans les pays en d6veloppement. References 1. Hendriksen CFM et al. The toxin binding inhibition test as a reliable in vitro alternative to the toxin neu- tralization test in mice for the estimation of tetanus antitoxin in human sera. Journal of biological stand- ardization, 1988,16: 287-297. 2. Hendriksen CFM, van der Gun JW, Kreeftenberg JG. Combined estimation of tetanus and diphtheria antitoxin in human sera by the in vitro toxin-binding inhibition (ToBI) test. Journal of biological standardi- zation, 1989, 17: 181-190. 3. Nikoletti S. Measurement of diphtheria and tetanus antitoxin in blood samples collected on filter paper disks. Epidemiology and infection, 1994, 112: 161- 170. 4. Christenson B, Bottiger M. Serological immunity to diphtheria in Sweden in 1978 and 1984. Scan- dinavian journal of infectious diseases, 1986, 18: 227-233. 5. Larsen K et al. The immunization in adults against diphtheria in Sweden. Journal of biological standardi- zation, 1987, 15: 109-116. 6. Mark A et al. Immunity and immunization of children against diphtheria in Sweden. European journal of clinical microbiology and infectious diseases, 1989, 8: 214-21 9. 7. Larsen K et al. The immunization of children with combined diphtheria and tetanus vaccine in Sweden. Joumal of biological standardization, 1987, 15: 103- 107. 8. Rumke HC et al. Serological evaluation of a simplified immunization schedule using quadruple DPT-polio vaccine in Burkina Faso. Vaccine, 1992, 11: 1113- 1118. 9. Booy R et al. Immunogenicity of combined diphtheria, tetanus and pertussis vaccine given at 2, 3 and 4 months versus 3, 5, and 9 months of age. Lancet, 1992, 339: 507-510. 10. Anderson EL et al. Clinical and serologic responses to acellular pertussis vaccine in infants and young children. American journal of diseases of children, 1987, 141: 949-953. 11. Anderson EL et al. Acellular pertussis vaccines in WHO Bulletin OMS. Vol 74 1996 281 H.A. Hong et al. infants: evaluation of single component and two-com- ponent products. Vaccine, 1994, 12: 28-31. 12. Edwards KM et al. Evaluation of a new highly purified pertussis vaccine in infants and children. Journal of infectious diseases, 1989, 160: 832-837. 13. Barkin RM, Samuelson JS, Gotlin LP. DTP reac- tions and serologic response with a reduced dose schedule. Joumal of pediatrics, 1984, 105: 189-194. 14. Hendriksen CFM et al. The use of the in vitro toxin binding inhibition (ToBI) test for the estimation of the potency of tetanus toxoid. Biologicals, 1991, 19: 23- 29. 15. WHO Expert Committee on Biological Standardiza- tion. Fortieth report. Geneva, World Health Organiza- tion, 1990: 87-179 (Annex 2) (WHO Technical Report Series, No. 800). 282 WHO Bulletin OMS. Vol 74 1996