Recombinant vaccinia viruses as live virus vectors for vaccine antigens: Memorandum from a WHO/USPHS/NIBSC Meeting*

B o Waan4 7 ( oranl nturns18 Bulletin ofthe World Health Organization, 63 (3): 471 -477 ( 1985) rcJ World Heaith Organization 1 985 Recombinant vaccinia viruses as live virus vectors for vaccine antigens: Memorandum from a WHO/USPHS/NIBSC Meeting* A scientific workshop sponsored by the World Health Organization, the US Public Health Service, and the National Institutefor Biological Standards and Control, London, was held in Bethesda, MD, USA, on 13 and 14 November 1984 to review progress in research relevant to the development of genetically and antigenically modified vaccinia viruses as live vaccines for human and veterinary use. The meeting was followed by an informal consultation convened by WHO to consider the advantages and disadvantages of this approach to vaccine design and production. The needs for further research and the potential role of WHO in coordinating and encouraging international activities in this area were discussed. This report summarizes the proceedings of the scientific workshop and the recommendations made by the WHO consultation. SUMMARY OF SCIENTIFIC SESSIONS The biological and molecular properties of vaccinia virus and its strategy of replication were reviewed in a scientific workshop held in Bethesda, MD, USA, on 13-14 November 1984. The origins of the different strains of vaccinia virus that had been used as vaccines against smallpox are uncertain but restriction endonuclease mapping has suggested that they represent a distinct and interrelated group of viruses. Although vaccinia virus replicates well on inoculation in a very wide range of mammalian hosts, sustained natural transmission in animals has not been reported. This would have considerable advantages for the use of the virus as a live vaccine vector. The ds DNA genome of vaccinia virus is very large with considerable variability within the terminal regions but with a high degree of conservation in the * This Memorandum was drafted by the signatories, listed on page 477, who were the participants in a WHO Consultation which followed the scientific workshop sponsored by WHO, the US Public Health Service (USPHS), and the National Institute for Biological Standards and Control, London (NIBSC). Requests for reprints should be addressed to the Division of Communicable Diseases, World Health Organization, 1211 Geneva 27, Switzerland. A French translation will appear in a later issue of the Bulletin. remainder of the molecule. There is a 28 kilobase region which is not essential for the replication of the virus. This region may be excised and replaced with genes coding for foreign microbial antigens.a b Genes have commonly been inserted into the thymidine kinase gene. The possibility of inserting foreign genes into vaccinia virus by recombinant DNA techniques forms the basis for the use of this virus as a live virus vector; the resulting vaccines could then be used for immunization against a variety of infectious agents of man and animals for which the genes coding for the protective antigen have been identified. Use of vaccinia virus as a vaccine The extensive experience gained with vaccinia virus vaccines in the WHO smallpox eradication programme was reviewed. Vaccinia vaccines have a number of features that were of paramount importance to the success of the WHO programme. These include the ease of preparation and high degree of stability, the low cost, the convenience of the route a PANICALI, D. & PAOLETTI, E. Proc. NatlAcad. Sci., 79:4927 (1982). bMACKETT, M. ET AL. Proc. NatlAcad. Sci., 79: 7415 (1982). 4543 -471- MEMORANDUM of administration, the easy visual assessment of 'take', and the long-enduring immunity following a single vaccine dose. These same features make vaccinia virus an attractive host in which to insert foreign genes for the development of new antimicrobial vaccines. The frequency of occurrence of adverse effects resulting from the extensive use of vaccinia in immunizations was discussed. Reactions involving the skin and central nervous system and the occurrence of generalized vaccinia as complications of vaccination were described. It was pointed out that there was some variation in these reactions between different strains of vaccinia virus. The level of complications also varied with the age of the recipient. The Lister and New York Board of Health strains were reported as having had the lowest frequency of complications. Almost all the vaccinia strains used routinely in vaccination programmes had been produced as calf lymph. Experience with vaccines that had been prepared in cell cultures was limited, the number of persons who had received such vaccines being small, so that no realistic comparison could be made of the relative frequencies of complications associated with cell-culture vaccines and conventional calf-lymph vaccines. If the same strain of vaccinia virus is used, there is no reason to expect differences in the frequency of complications associated with cell culture or calf lymph vaccines. It was emphasized that important information on the characteristics of modi-fied vaccinia viruses could be derived from clinical studies involving small numbers of intensively studied subjects by careful measurement of virus replication and the various parameters of the immune responses to vaccination. Moreover, studies of the virulence of the modified vaccine strains for different species of animals might be of value in predicting safety in man. Some data on the genetic basis of attenuation with ectromelia, vaccinia, thymidine-kinase defective (TK-) vaccinia mutants and recombinant vaccinia viruses, using mice as the host species, were presented. Nevertheless, the ultimate evidence of safety and efficacy of any new modified vaccinia vaccine intended for human use would depend on studies in man. Any veterinary vaccine would need to be studied extensively in the intended recipient species. Production of vaccine virus Methods for the production of vaccinia vaccines were discussed. Conventionally the virus had been grown on the skin of calves. This can be done cheaply and rapidly, about 106 human doses of vaccine being obtained from one production animal. However, this method does not invariably yield a bacteria-free preparation. Considerable doubts were expressed by participants as to whether vaccines prepared by this traditional method would now be acceptable. Vaccinia could also be produced inexpensively in tissue culture; this was seen as a more generally acceptable method for production. The considerable technical problems associated with vaccine production by tissue-culture methods in many countries in the developing world were discussed. It was suggested that, if necessary, technical assistance might be made available for the transfer of tissue- culture technology to developing countries. Insertion offoreign genes into vaccinia virus for use as potential vaccines The successful incorporation of genes coding for antigens relevant to protective immunity, including antigens of hepatitis B, herpes simplex, influenza, rabies and Epstein-Barr viruses of man, and vesicular stomatitis and transmissible gastroenteritis viruses of animals, as well as antigens of protozoan parasites (malaria), was reported. With only one exception, a single inoculation of the recombinant virus induced the formation of antibody against the foreign antigen. The exception was when vaccinia virus containing the hepatitis B virus surface antigen gene was used in chimpanzees; a secondary response was obtained after the animals were challenged with hepatitis B virus but the animals were protected against clinical disease, although infection did occur as shown by the occurrence of core antigen. When DNA coding for the haemagglutinin (HA) or the nucleoprotein (NP) of influenza A virus was inserted into vaccinia virus and the recombinant vaccine was inoculated into mice, the animals were shown to be primed for specific cytotoxic-T cell (Tc) responses. On subsequent challenge with either influenza virus or the recombinant vaccinia virus and using cells infected with influenza A virus as targets, secondary Tc responses were obtained. The Tc responses in animals infected with the recombinant virus possessing the gene for NP were broadly cross- reactive for influenza A virus of different subtypes. In contrast, the Tc response to the gene for HA was predominantly subtype-specific. Intranasal adminis- tration in mice of vaccinia possessing the gene for influenza HA induced specific IgA production in the upper respiratory tract as well as a systemic IgG response. In contrast, intradermal inoculation elicited only the latter response. Either route of vaccination solidly protected the mice against lower respiratory infection with the homologous influenza virus, but only the intranasal route provided protection against upper respiratory infection. These findings clearly illustrate the value of vaccinia virus recombinants in the experimental analysis of the 472 RECOMBINANT VACCINIA VIRUS VACCINES immune response. For example, recognition of at least two antigens, HA and NP, by effector-T cells may be important in recovery from infection, and antibody against HA may be required to prevent infection. It was also concluded from these studies that, for viruses that initiate infection at mucosal surfaces, the route of administration of the vaccinia vaccine might be of critical importance. Vaccinia virus recombinants that expressed a surface glycoprotein (gD) of herpes simplex virus type I elicited a neutralizing antibody response in mice, but did not induce a primary Tc response or prime for a secondary Tc response. A single intradermal inoculation protected the animals against lethal challenge with both types 1 and 2 herpes simplex virus as well as against the establishment of latency in the trigeminal nerve. A vaccinia recombinant that expressed the gene coding for the large membrane protein of Epstein-Barr virus was also described. Successful experiments were also reported with a vaccinia virus recombinant containing the gene coding for the surface glycoprotein (G) of rabies virus. Mice receiving 104 pock-forming units of the recombinant virus produced a high-titre neutralizing antibody response and solid immunity to intra- cerebral challenge with Street virus. Moreover, immunization of animals was successfully achieved using G protein in inactivated harvests of cell cultures infected with the modified vaccinia virus. The yield of G protein from tissue-culture cells infected with the rabies-vaccinia virus recombinant was reported to be tenfold higher than that produced in cells infected with rabies virus by itself. The live recombinant virus efficiently primed mice to give a secondary Tc response using rabies-infected target cells. Approaches to the development of a number of veterinary vaccines were described. A vaccinia recombinant that expressed the glycoprotein antigen of vesicular stomatitis virus partially protected cattle against intralingual challenge. In addition, a recombinant vaccinia virus containing a gene coding for a surface glycoprotein of transmissible gastro- enteritis virus of pigs -this is a coronavirus -elicited neutralizing antibody responses in rats. More work needs to be done in defining Tc responses to recombinant virus preparations and particularly to the inability of any recombinant virus preparation so far studied to elicit a primary Tc response. Considerations ofsafety and efficacy In discussions it was agreed that, if progress is to be made concerning the potential safety and efficacy of new vaccines based on genetically modified vaccinia virus, there would need to be much more research work directly relevant to public health issues. Further studies were required on potential adverse effects of "wild type" and recombinant vaccinia vaccines and the possibility of viral transmission within or between species from infected hosts. Genotypic and phenotypic markers of attenuation would need to be identified more precisely. Concerning the protective efficacy of recombinant vaccine strains, the animal studies so far reported are promising but the importance of the route of inoculation was stressed as well as the details of the construction of the recombinants. It was emphasized that the vaccinia system would not be relevant for protection against diseases where the important immunizing antigen consists of lipid or carbohydrate but only where protein or glycoprotein antigens are involved. Thus far, hybrid vaccinia viruses have been shown to protect against systemic infections by viruses as diverse as hepatitis B, rabies and herpes simplex viruses and against lower respiratory infection by influenza virus. However, whether vaccinia virus hybrids would protect against infections of the gastrointestinal tract has not yet been experimentally evaluated. Indeed, dermal inoculation might not provide the local immunity in the gut that is required to protect against enteric infections. For this reason, potential vaccine vectors other than vaccinia virus should be considered. It was strongly emphasized that the anatomical site of replication of any vector, i.e., skin, gastrointestinal tract or respiratory tract, may be of importance in determining the type of immune response stimulated by vaccination. Selection of a vector producing an immune response relevant to protection against the target disease may thus be an important factor in developing an effective vaccine. It seems likely that the first field trials with recombinant vaccinia viruses will be in the veterinary field. However, concern was expressed about the possible transmission of vaccinia from animals to man. It is clear that careful studies would need to be done under well-controlled conditions for each recombinant intended for use as vaccine to seek evidence of such transmission. Alternative vectors Although the workshop was concerned primarily with genetically-modified vaccinia viruses as potential live vaccines, the group concluded that research work on other microbial vectors should continue. The feasibility of using herpes simplex virus as a vector was described, e.g., the construction of herpes simplex virus recombinants with considerably attenuated virulence but an unchanged ability to replicate and to protect against challenge by the wild type virus. 473 MEMORANDUM CONCLUSIONS AND RECOMMENDATIONS The participants in the WHO consultation that followed the workshop in Bethesda considered the implication of the scientific developments under five headings: (1) The need for further molecular studies relevant to vaccine efficacy The efficacy of vaccinia virus recombinants as vaccines will depend on their ability to elicit a protective immunological response. The quantity of antigen that must be expressed by a recombinant vaccinia virus vector in order to confer immunity will vary depending on the immunogenicity of the protein and the degree and duration of protection required. A thorough understanding of the control of vaccinia virus gene expression is thus an important requisite for development of effective vaccines. There is an important need to encourage and support research in this area. Specifically, the nature of transcriptional regulatory signals, transcriptional termination sequences, and translational regulation should be determined. Because of the molecular complexity of poxviruses and the few laboratories working in this area, only limited information on this subject is currently available. Further work is also needed to determine the best ways of linking the foreign gene to vaccinia virus transcriptional regulatory sequences, the optimal sites of foreign gene insertion, and the expression of multiple foreign genes. The construction of suitable recombinant vaccinia strains depends on the availability of cloned genes of the target pathogen that encode the microbial antigens responsible for the protective humoral and cell-mediated immune response. In many cases, how- ever, the appropriate genes are not yet known and in others they have not yet been cloned. The deficiency in this area is particularly acute with protozoan, helminthic and bacterial diseases but also exists for most virus diseases of medical and veterinary importance. There is a clear need to encourage research on the identification and cloning of genes that express protective antigens. (2) The need to establish the level of attenuation of the recombinant vaccinia viruses At present, there is no generally accepted laboratory marker to predict virulence or attenuation of vaccinia virus for man. However, studies of mortality or pathogenesis in infected mice have been found of value. Alterations in the vaccinia virus genome that are concomitant with the selection of recombinants may alter the virulence of the virus. It has been noted that interruption of the TK gene of the virus by insertion of foreign DNA reduced its virulence in mice. It is important to extend further the studies of virulence and attenuation of vaccinia virus, as well as recombinants derived from it, by using other animal species besides mice. These should include primates and species such as cattle, which might be targets for immunization with recombinant vaccines. As required for vaccination with any live virus, it is essential to study the attenuation of the recombinant virus, not only in vaccine recipients but also after several passages of the virus in the same species. Development of "faithful" genetic markers for attenuation would greatly assist these studies. Since potential vaccine recipients may include immune- suppressed individuals, the attenuated strains of vaccinia should be compared for their capacity to induce disease in immunosuppressed hosts as well as in immunologically-competent hosts. Since successful immunization with recombinant vaccinia or other vectors may require different routes of injection, careful consideration should be given to the study of recombinants using various routes of administration of the virus. It has been shown, for example, that parental vaccinia virus, before insertion of another viral gene, is neurovirulent in certain animals by intraperitoneal injection but not by other routes of inoculation. Insertion of antigen- coding regions of herpes simplex virus or hepatitis B virus genomes was found to render the recombinant virus avirulent after intraperitoneal injection, possibly because in most constructs the foreign gene was inserted into the TK gene. Studies aimed at the further attenuation of vaccinia virus should be concentrated on one (or a very few) carefully selected, well documented reference strains. Preference should be given to strains that have been used extensively in the past for effective human immunization with a minimum of side-effects. Effort should be directed at finding genetic markers that could distinguish the reference strains from others. Such reference virus(es) should be used as controls in biological safety and efficacy studies of any recombinant intended for vaccine use; for example, in studies of virulence, animal host range, and vaccinia- specific immune responses. (3) Appropriate model systems for safety and efficacy testing Vaccinia virus has a broad host range but the clinical manifestations of infection vary widely and unpredictably from species to species. Safety and efficacy data obtained in one animal species cannot therefore be assumed to be applicable to another species. 474 RECOMBINANT VACCINIA VIRUS VACCINES In the past, laboratory mice have been used for assessing the virulence and efficacy of vaccinia virus, and standardized protocols using defined, inbred, SPF (specific pathogen-free) mice for these tests should be developed. Information should be collected on the relationship of virulence in mice of recombinant vaccinia viruses with their virulence in the species for which the vaccine is intended. Alternative test systems should be developed if the results in mice do not correlate well with the results observed in the vaccinated animal species or in man. Changes in host range or tissue tropism of vaccinia viruses may occur as a result of their genetic modification and could also be caused by changes in the virus envelope through incorporation of the products of the foreign viral genes into the vaccinia virus genome. Before field studies in animals or man are performed using any recombinant vaccinia virus, the latter should be exhaustively tested in cell-culture systems and in experimental animals under strict isolation to determine if such changes have occurred. It is of critical importance that shedding of the recombinant vaccinia virus and its ability to transmit naturally in animals or man, or between species, be determined. The genetic stability of the vaccine virus during repeated passage in animals or man is of particular importance and needs to be fully evaluated. Recombinant vaccinia viruses intended for use in veterinary vaccines, following their initial study in small laboratory animals, should be tested for safety and efficacy in the animal species for which the vaccine is intended. Vaccinia viruses intended for use in man must first be tested in laboratory animals, preferably including non-human primates. For each recombinant strain, molecular or biological markers for virulence and immunization efficacy should be firmly established. It must be emphasized, however, that the rare CNS complications (occurring in a few cases per 106 recipients for vaccinia vaccines) can be evaluated only in large-scale trials in man and that the risk must be carefully considered for each potential vaccine recombinant. Testing for immunogenicity in mice, and, where appropriate, in other animal species should evaluate humoral as well as cell-mediated immune responses. It should include determination of the immuno- globulin classes of the specific antibodies induced by the vaccine and identification of the classes of T lymphocytes stimulated. Efficacy should be determined by experimental challenge studies in animals or by natural exposure in controlled field trials in the species in which the vaccine will be used. The persistence of immunity will need to be determined and the efficacy of repeated booster vaccinations should be evaluated. (4) Advantages of vaccinia-derived recombinant vaccines and priorities for vaccine development The use of vaccinia virus as a live vaccine vector has a number of potential advantages for immunization against diseases that are of worldwide public health priority, provided that the recombinant strains retain the desirable features of the parental vaccinia vaccines. The properties that make vaccinia virus a potentially useful vehicle for vaccine development are: -low cost: the major costs relate to research and development; production costs should be minimal and constant, regardless of the number of antigens expressed by the recombinant virus. -ease of administration: needles and syringes will not be required. -vaccine stability: recombinant vaccinia virus is assumed to be highly stable with a long shelf-life; the field use of the vaccine should be minimally dependent on a "cold chain". -polyvalency: the use of recombinant vaccines offers the possibility of vaccination against several diseases in a single dose of vaccine; the public health value of the vaccine would potentially increase with the number of specific antigens incorporated into a single vaccine strain. -known adverse reactions: the known adverse reactions with vaccinia virus vaccines used previously are well documented; their incidence must be weighed against the known adverse reactions associated with the conventional existing vaccine which a new recombinant vaccine might replace. -potential new vaccines: the use of recombinant vaccine is particularly promising for immunization against agents for which vaccines are not at present available or which are difficult to produce. Replacement of an existing vaccine would be favoured if more than one antigen could be incor- porated in the recombinant vaccine, or where the present vaccine is expensive to produce or administer or is unsafe or ineffective. Production of the recombinant vaccinia vaccine in cell cultures would be desirable. However, vaccine produced on animal skin may in some cases be shown to be acceptable. Vaccinia virus produced in tissue culture will involve more advanced technology than that previously required for production of calf-lymph vaccine. 475 MEMORANDUM (5) Recommended action for the eventual formulation of control requirements for the new vaccines The requirements for controlling human vaccines utilizing vaccinia virus as a vector for heterologous antigens will be developed progressively as more information relating to attenuation, production, and clinical safety and efficacy become available. These requirements will undoubtedly have to take into account the problems connected with this particular approach to vaccine development as well as issues pertinent to the specific vaccine. Alternative, improved methods of inserting genes into vaccinia virus are likely to be developed in the future and experimental vaccines involving antigens of numer- ous infectious agents will be studied. Each new cloning technique and each new vaccinia virus- antigen recombinant may necessitate new or modified requirements. The requirements should be addressed to issues pertinent to live virus vaccines in general, such as the characteristics of the seed virus, use of a seed-lot system, description of the substrate, and methods for production and safety, potency, identity and stability testing. Issues specifically relevant to seed viruses for recombinant vaccinia virus vaccines will probably include a description of the parent vaccine virus, methods for gene cloning, confirmation of the genome structure of the seed virus obtained, demonstration of the genetic stability of the seed virus, description of the behaviour of the virus in animal models, and a summary of the clinical trials that demonstrate that the seed strain is suitable for vaccine production. Unique types of control testing on vaccine lots may include verification of genome structure and antigen product as part of identity testing, specific safety tests in animal models, assessment of levels of antigen expression in infected cells as a potency measure, and other factors. The eventual requirements should be based on the results of extensive laboratory and clinical research. The establishment of one or more collaborating centres for research on microbial vectors for use as vaccines should be considered. The proposed terms of reference of the first centre, the establishment of which should be given immediate priority, are: - to act as a repository for information on the use of vaccinia virus as a vector for vaccine; -to act as a repository for plasmid vectors and vaccinia virus strains to be used for construction of hybrid viruses; -to advise WHO on the need for scientific meetings relevant to vaccines based on genetically- modified microbial vectors; -to assist WHO in the planning and performance of collaborative studies on pertinent issues; -to advise WHO on research needs. In addition, the establishment of a second collaborating centre based in a vaccine production laboratory should be considered. Recommendations (1) There is an important need to support and encourage research work on the regulation and enhancement of expression of foreign genes by recombinant vaccinia virus, or other potential micro- bial vectors. The role of international coordination in this field should be considered. (2) Progress in vaccine development involving recombinant DNA techniques is dependent on the identification of microbial proteins involved in the protective immune response and the characterization of the genes which encode them. Research in this area should be encouraged. There is a particular need for studies on bacteria, protozoal and helminthic parasites and on certain groups of viruses for which effective vaccines are not available. (3) Although most progress has been made with vaccinia virus as a potential vector for the development of live vaccines, the group considered that research on the other vectors should be encouraged. Candidate vectors may include bacterial species, such as those capable of colonizing the gut, as well as viruses other than vaccinia. (4) The exchange of information and the establishment of technical collaboration between laboratories working in the above areas should be promoted. (5) It is likely that experience with the use of live vaccines employing vaccinia as a virus vector will first be gained in the veterinary field. Studies in animals will be relevant to an understanding of the safety and efficacy of similar vaccines fot use in man. It is recommended that appropriate steps should be taken for close liaison with veterinary authorities to ensure as far as possible that the veterinary use of such vaccines should be accompanied by rigorous surveillance to detect the possible spread of vaccine virus between animals and also to man. In particular, WHO should seek to collaborate with FAO on research and the development of new vaccinia-derived vaccines for veterinary use. (6) It is impOrtant that as much experience as possible be obtained with one or a few strains of vaccinia virus that could be designated as reference strains for laboratory research and for vaccine studies. The establishment of reference virus strains will be essential to provide a basis for assessment and 476 RECOMBINANT VACCINIA VIRUS VACCINES 477 comparison of any new recombinant vaccinia virus for use as a vaccine. WHO has an important role in fostering collaborative work on the establishment of these reference strains. (7) In the selection of a reference strain of vaccinia virus for use in vaccine development, preference should be given to strains that had been used extensively in the past for the preparation of vaccines with a proven record of safety and efficacy, and whose molecular and genetic features are well known. A group of scientists should be formed to advise on the evaluation and selection of suitable reference strains. (8) Consideration should be given to the designation of one or more WHO collaborating centres for research on microbial vectors for use in vaccines. (9) Every encouragement should be given to the strengthening of national laboratories involved in the standardization and control of vaccines. Transfer of technology for the production of new vaccines based on recombinant vaccinia or other vectors should be accompanied by the establishment of adequate facilities for the standardization and control of the new vaccines., This would involve the use of appropriate techniques in molecular biology as well as conventional biological methods. (10) With advice from appropriate scientists, guidelines and, eventually, the requirements for new recombinant vaccines based on vaccinia or other vectors, should be formulated. * * G. L. Ada, Microbiology Department, The John Curtin School of Medical Research, Canberra City, Australia (Chairman and Joint Rapporteur) F. Brown, Wellcome Biotechnology Ltd., Pirbright, Woking, Surrey, England (Joint Rapporteur) F. Deinhardt, Max. v. Pettenkofer Institute, Munich, Federal Republic of Germany W. Dowdle, Centers for Disease Control, Atlanta, GA, USA F. Filatov, Institute of Virology, Academy of Medical Sciences, Moscow, USSR A. C. Hekker, State Institute of Public Health, Bilthoven, Netherlands D. A. Henderson, School of Hygiene and Public Health, The Johns Hopkins University, Baltimore, MD, USA W. S. Jordan, Microbiology and Infectious Diseases Programme, National Institutes of Health, Bethesda, MD, USA D. T. Karzon, Department of Pediatrics, School of Medicine, Vanderbilt University, Nashville, TN, USA S. Kato, Research Institute for Microbial Diseases, Osaka University, Osaka, Japan H. Koprowski, Wistar Institute, Philadelphia, PA, USA B. Moss, Macromolecular Biology Section, National Institute of Allergy and Infectious Diseases, Bethesda, MD, USA G. V. Quinnan, Bureau of Biologics, Bethesda, MD, USA G. C. Schild, National Institute of Biological Standards and Control, London, England (Joint Rapporteur) K. S. Warren, The Rockefeller Foundation, New York, NY, USA WHO Secretariat I. Arita, Smallpox Eradication, WHO, Geneva, Switzerland F. Assaad, Division of Communicable Diseases, WHO, Geneva, Switzerland R. Henderson, Expanded Programme on Immuniz- ation, WHO Geneva, Switzerland F. Pineiro, Communicable Diseases, WHO Regional Office for the Americas, Washington, DC, USA G. Torrigiani, Immunology, WHO, Geneva, Switzerland