Cover Image

HARDBACK
$52.95



View/Hide Left Panel

2
Progress in Vaccine Development

In 1985 the Institute of Medicine (IOM) published its initial report on vaccine development: New Vaccine Development, Establishing Priorities, Vol. 1, Diseases of Importance in the United States (IOM, 1985a). This was the first report of a two-part study (IOM, 1985a,b) conducted by the Committee on Issues and Priorities for New Vaccine Development at the request of the National Institute of Allergy and Infectious Diseases (NIAID) of the National Institutes of Health (NIH). NIAID contracted with IOM for assistance in developing an approach to setting priorities for accelerated vaccine development. Committee membership embraced expertise in all those fields judged to be important to such a study, and therefore, in addition to research scientists and physicians, the committee also included economists, sociologists, ethicists, epidemiologists, industry leaders, and public health experts. As part of its charge, the committee was requested to develop a decisionmaking framework for selecting candidate vaccines of importance to the U.S. population and to use such a system to rank these candidate vaccines. A second component of the study involved prioritization of the vaccines needed by technologically less developed nations. That part of the study also modified the model to rank potential vaccines for use on an international basis. These findings were published in a second volume of the committee’s report (IOM, 1985b).

The committee’s analysis was based on a quantitative model in which vaccine candidates were ranked according to two principal characteristics: expected health benefits (reduction of morbidity and mortality) and expected net savings of health care resources. To compare the health impacts of disease and the potential benefits of vaccines, a measurement system based on units of infant mortality equivalents was used. The best available data were used, but where the information was incomplete, estimates and judgments by experts were used. The committee adopted a format that was flexible so that new candidate vaccines could be assessed similarly or current candidates vaccines could be reassessed with new data. A final



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 17
Vaccines for the 21st Century: A Tool for Decisionmaking 2 Progress in Vaccine Development In 1985 the Institute of Medicine (IOM) published its initial report on vaccine development: New Vaccine Development, Establishing Priorities, Vol. 1, Diseases of Importance in the United States (IOM, 1985a). This was the first report of a two-part study (IOM, 1985a,b) conducted by the Committee on Issues and Priorities for New Vaccine Development at the request of the National Institute of Allergy and Infectious Diseases (NIAID) of the National Institutes of Health (NIH). NIAID contracted with IOM for assistance in developing an approach to setting priorities for accelerated vaccine development. Committee membership embraced expertise in all those fields judged to be important to such a study, and therefore, in addition to research scientists and physicians, the committee also included economists, sociologists, ethicists, epidemiologists, industry leaders, and public health experts. As part of its charge, the committee was requested to develop a decisionmaking framework for selecting candidate vaccines of importance to the U.S. population and to use such a system to rank these candidate vaccines. A second component of the study involved prioritization of the vaccines needed by technologically less developed nations. That part of the study also modified the model to rank potential vaccines for use on an international basis. These findings were published in a second volume of the committee’s report (IOM, 1985b). The committee’s analysis was based on a quantitative model in which vaccine candidates were ranked according to two principal characteristics: expected health benefits (reduction of morbidity and mortality) and expected net savings of health care resources. To compare the health impacts of disease and the potential benefits of vaccines, a measurement system based on units of infant mortality equivalents was used. The best available data were used, but where the information was incomplete, estimates and judgments by experts were used. The committee adopted a format that was flexible so that new candidate vaccines could be assessed similarly or current candidates vaccines could be reassessed with new data. A final

OCR for page 17
Vaccines for the 21st Century: A Tool for Decisionmaking recommendation urged NIAID and other responsible agencies to improve the epidemiological data that are used to compare diseases. The variable quality of the data in some areas and the total absence of data in others were serious impediments to the development of a comprehensive prioritization scheme. PRIORITIES OF THE IOM COMMITTEE IN 1985 Table 2–1 presents the 1985 committee’s final list of pathogens for which vaccines were analyzed. For several pathogens two different forms of vaccine were considered (e.g., attenuated live virus or glycoprotein). By varying different assumptions (e.g., utilization, discount rates), vaccines against the first five pathogens retained highest priority. The committee stated that an improved pertussis vaccine (acellular antigens) merited unique treatment because of its potential for restoring public confidence in all immunization programs. Indeed, since the report was issued, a number of acellular pertussis vaccines and new combinations of acellular pertussis vaccines with other antigens have been licensed in the United States. Initially, acellular pertussis vaccines were recommended only for the fourth and fifth doses of the childhood immunization schedule, but acellular pertussis vaccines are now recommended for all doses in the immunization schedule beginning at 2 months of age. Their diminished local reactivity and systemic manifestations have rendered them highly acceptable to both health care providers and parents. This has greatly restored confidence in childhood immunizations, which had been eroded previously by concerns about whole-cell pertussis vaccines. The hepatitis B virus (HBV) vaccine, prepared by recombinant technology, has also been licensed for use and is recommended and incorporated into the infant immunization schedule. Additionally, it is widely used for various high-risk adult populations including health care workers. Conjugate vaccines against Haemophilus influenzae type b (Hib) have been the remarkable success of the past 8 years. With their widespread use in infancy, the annual rates of invasive disease caused by this organism among U.S. children under 5 years of age have been reduced from 40 to 1 per 100,000 population. Varicella-zoster virus vaccine was licensed in the spring of 1995 and is recommended for infants at 1 year of age. Physicians in practice are beginning to use the vaccine, and a number of states provide it in their public health programs. Of the remaining candidate vaccines included in the 1985 analysis, only hepatitis A virus (HAV) vaccine (1995) and rotavirus vaccine (1998) have been licensed. Vaccination against HAV has been recommended for groups at special risk. Over half of the candidate vaccines analyzed in the 1985 report on vaccine priorities have not yet been licensed. Several are still 15 years from licensure, in this committee’s opinion. Although the committee did not analyze in depth each unlicensed candidate vaccine from 1985, obvious factors hindering progress

OCR for page 17
Vaccines for the 21st Century: A Tool for Decisionmaking TABLE 2–1 Status of Vaccines Prioritized in 1985 Pathogen Status Hepatitis B virus Recombinant hepatitis B virus surface antigen licensed and in widespread use. Respiratory syncytial virus Purified fusion (F) protein in phase II studies. Haemophilus influenzae type b Glycoconjugates licensed and in widespread use. Influenza virus A and B Cold-adapted live, attenuated virus vaccine in phase III trials; baculovirus-expressed recombinant HA subunit in phase III trials. Varicella-zoster virus Live attenuated virus licensed and in use. Group B streptococcus Glycoconjugates of five serotypes in phase II studies. Parainfluenza viruses Cold-adapted live, attenuated type 3 virus in phase II trials; bovine live, attenuated virus in phase II trials. Cytomegalovirus Glycoprotein subunit in phase I studies. Rotaviruses Attenuated live human-rhesus and bovine-human reassortants in phase III trials at beginning of project. Licensure in 1998 of one product. Neisseria gonorrhoeae Early basic research on various proteins. Hepatitis A virus Inactivated HAV particles licensed and in early use. Coccidioides immitis Formalin-killed spherules in phase III clinical trials. Herpes simplex 1 and 2 Glycoprotein (D) recombinant of type 2 in phase II trials; attenuated recombinants in phase I studies. Bordetella pertussis Acellular products (DTaP) licensed and in widespread use. toward licensure include unexpected obstacles in research progress (either in the more basic research phases, in clinical trials, or in scale-up processes for development and production). Alternatively, steady progress could have been made but at a much slower pace than expected. Slow progress could be attributed to either lack of scientific interest on the part of researchers or lack of adequate funding for the R&D. As discussed in Chapter 7 of the report, inadequate interest on the part of funders, such as private vaccine R&D companies, can reflect concerns about profitability because of either small market potential or possible costs due to liability for adverse events. The committee had been told in early discussions with vaccine researchers that a primary weakness of the 1985 report was that it was overly optimistic about how long it would take for licensure of several candidates. The present analysis includes a longer time frame than the 1985 effort. Models such as that

OCR for page 17
Vaccines for the 21st Century: A Tool for Decisionmaking provided in 1985 and in this report do not predict which vaccines will be developed within a specific time frame. They provide comparative cost-effectiveness analyses for candidate vaccines that the committee assumes could be developed and licensed within a specific time frame. The development of respiratory syncytial virus (RSV) vaccines has been slower than anticipated, in large part because of the extraordinarily cautious approach to the implementation of clinical trials because of the unfortunate experiences with the early inactivated RSV vaccines. The inactivated vaccines augmented clinical disease and resulted in increased rates of hospitalizations and some deaths when administered to the youngest infants, the group at highest risk from RSV disease. Clinical trials with both attenuated live virus and surface glycoprotein RSV vaccines are under way. Influenza virus vaccines consisting of both attenuated live virus variants and a number of subunit preparations are also under continuing research and development. In contrast, the likelihood of successful licensure of parainfluenza virus vaccines, cytomegalovirus (CMV) vaccine, Neisseria gonorrhoeae vaccine, and Coccidioides immitis vaccines remain more remote. Studies of conjugate group B streptococcal (GBS) vaccines have been promising, particularly in immunization of late third-trimester pregnant women for the prevention of neonatal invasive GBS disease. However, pharmaceutical firms appear to have little enthusiasm for investing in the production of such a vaccine. Concerns regarding litigation that might ensue following unfavorable pregnancy outcomes (discussed in the following section) remain the most visible obstacle. Table 2–1 presents the current stage of development of vaccines against those pathogens assessed in 1985. LITIGATION AS A BARRIER TO VACCINE DEVELOPMENT The mid-1980s was a time of great struggle and even crisis for the development and manufacture of vaccines and vaccination program implementation. Public trust in vaccines was shaken and litigation concerns caused several major manufacturers to reduce or eliminate their vaccine development programs. The National Childhood Vaccine Injury Act (NCVIA; P.L. 99–660) was enacted in 1986. The legislation established a compensation fund for people who suffered specific serious adverse health effects that could potentially be attributed to vaccination with mandatory childhood vaccines (diphtheria and tetanus toxoids and pertussis vaccine, measles-mumps-rubella vaccine, oral polio vaccine, inactivated polio vaccine, and individual antigens within those vaccines). IOM embarked on two major projects to evaluate the medical and scientific literature regarding the causal association between vaccines and adverse events (IOM, 1991, 1994a,b). That body of work has been used by the U.S. Department of Health and Human Services to evaluate and refine the conditions and circumstances warranting compensation by the program. The Vaccine Injury Compen-

OCR for page 17
Vaccines for the 21st Century: A Tool for Decisionmaking sation Program (NVICP) is funded by excise taxes on the sale of vaccines covered by the program. Although NCVIA does not protect vaccine manufacturers from all litigation, the number of suits filed against manufacturers greatly decreased, public trust in vaccines increased, and vaccine research and development returned to its previous levels. Serious litigation issues that have not been ameliorated by the 1986 legislation remain, however. Vaccines that are not mandated for use in children are not covered by NVICP. Federal vaccine advisory bodies have discussed the possibility of including influenza and pneumococcal vaccines in the program, which are aimed primarily at adults. Advocacy organizations key to the passage of the 1986 law are not supportive of that move, and no system of protection against litigation exists for these vaccines. It must be noted, however, that with the exception of year-specific concerns about a relation between influenza vaccine and Guillain-Barré syndrome, these vaccines have not caused safety concerns similar to those engendered by the childhood vaccines. This difference is probably due to differences in host reactions as well as the fact that childhood vaccines are, for the most part, mandated by the Federal government while adult vaccines are not (IOM, 1997b). Immunization of Pregnant Women Many researchers believe that litigation concern with regard to the immunization of pregnant women is a key factor in the slow progress in the development of a vaccine against GBS in particular. A discussion of the principles and problems surrounding immunization of pregnant women is necessary to understanding this barrier to vaccine development, which has not been resolved since the 1985 report. Preventing infections in neonates and young infants by vaccinating pregnant women is a concept that is more than a century old. The rationale for immunization of pregnant women is based on two premises: (1) the occurrence of a substantial burden of disease from certain infectious agents during the first 3 to 6 months of life, an interval before which protection by active immunization of the infant could be achieved, and (2) the likelihood that immunization prior to or during pregnancy would induce high concentrations of specific antibodies in maternal serum that would be transferred across the placenta, thereby passively protecting the infant. This strategy has been studied extensively for the prevention of diphtheria, tetanus, and pertussis infections in infants. Worldwide, immunization of pregnant women for the prevention of infection in the infant is used routinely for neonatal tetanus, a major cause of infant mortality (Schofield, 1986). The World Health Organization (WHO) has recommended that this approach be extended for the prevention of diseases caused by other infectious agents.

OCR for page 17
Vaccines for the 21st Century: A Tool for Decisionmaking The feasibility of immunizing pregnant women for the prevention of several infectious diseases whose contemporary burdens in young infants are substantial has been extensively investigated. The vaccines studied include those consisting of antigens from bacterial pathogens (GBS [Baker et al., 1988], Hib [Mulholland et al., 1996], Neisseria meningitidis groups A and C, Streptococcus pneumoniae [McCormick et al., 1980; O’Dempsey et al., 1996] and viral pathogens (HBV, influenza virus, rabies, RSV, yellow fever virus), and others have been proposed for study (CMV, herpes simplex virus, human immunodeficiency virus, rotavirus). Immunization of pregnant women is an approach that can avoid the obstacles presented by the immunologic immaturity of the neonate, has the potential to defer the need for active immunization of the infant or to decrease the number of doses needed to achieve protection during the first year of life, and may prevent the transmission of infection from the mother to the neonate. Despite considerable scientific data underscoring the feasibility, safety, and cost-effectiveness of this approach, progress in the United States has been slow or wanting. Obstacles to this approach are said to involve ethical, legal, and sociological concerns, but issues concerning the liability of vaccine manufacturers dominate the others (Insel et al., 1994; Linder and Ohel, 1994). Scientific evidence pales when concern over the liability of vaccine manufacturers arises, and vaccine manufacturers appeal to the government for indemnification before they pursue studies of existing vaccines or the development of additional reagents appropriate for immunizing pregnant women. The committee reviewed immunization of pregnant women as a vaccination strategy in some detail, believing it to be scientifically valid for use in the United States for preventing several infectious diseases in young infants. However, this approach deserves special consideration because of the obstacles mentioned above. Thus, it is appropriate to underscore the rationale for this approach. Alternatives are discussed and critiqued in Chapter 7. Passive immunization of the neonate and young infant depends on selective placental transfer of plasma proteins, and the transfer of immunoglobulins is limited to immunoglobulin G (IgG). Passage of antibodies is both passive (directly proportional to the maternal serum IgG concentration) and active (binding of IgG to Fc receptors, followed by receptor-mediated endocytosis). The latter accounts for the preferential transport of IgG 1 and IgG3, which have greater affinities than IgG2 for binding to Fc receptors. Passage of antibodies begins at 8 weeks of gestation, but the level remains low until about 20 weeks. Neonatal cord serum IgG levels correlate with gestational age. At 32 weeks of gestation cord serum has an IgG level approximately half of that at 40 weeks, when the levels are equal to or somewhat greater than those in maternal serum. Maternal IgG has a half-life of 3 to 4 weeks, but since the duration of passive antibody protection is dictated by the actual level at birth, protection may last for 3 to 6 months. The immune response of pregnant women is similar to that of nonpregnant women (Halsey and Klein, 1990), but there is a delay between the time of immunization and the time of fetal acquisition of maternal antibodies. This is the interval required for the mother to generate an IgG antibody response and for the

OCR for page 17
Vaccines for the 21st Century: A Tool for Decisionmaking antibodies to equilibrate with the circulating IgG pool and be transported across the placenta. For tetanus toxoid the optimal timing for immunization of a pregnant woman to achieve neonatal protection is 60 days before delivery (or at least 20 days after administration of the second dose) (Chen et al., 1983). The ideal vaccine should induce high maternal levels of IgGl antibodies; the maximum response should occur after the administration of one dose, reaching its peak within 2 weeks of immunization; and protective levels should persist for several years, providing protection in subsequent pregnancies. For most vaccines, the vaccine would be given early in the third trimester (28 to 32 weeks of gestation), a time when organogenesis is complete and when most events associated with adverse pregnancy outcomes are past. Theoretically, this timing would also provide protection for many prematurely born infants. For some diseases for which immunization of the pregnant woman has been proposed (e.g., neonatal and pregnancy-related GBS infections), active immunization of the infant would be unnecessary because susceptibility is limited to young infants, pregnant women, and adults with either defined underlying medical conditions or advanced age. For others, such as Hib and RSV infections, active immunization would also be required. Maternal antibody can interfere with infant responses to live-virus vaccines, but this has not been demonstrated for inactivated viral vaccines, such as influenza virus vaccine, or for bacterial antigens, including tetanus toxoid, serogroup A and C meningococcal polysaccharides, or Hib polysaccharide (Insel et al., 1994). While the issue of suppression, activation or priming, or alteration of the repertoire of antibody responses in the infant should be studied as vaccines are developed, the evidence to date is reassuring. A CASE STUDY OF SUCCESS As mentioned above, several of the vaccines recommended for accelerated development in the 1985 IOM report have been licensed for use, and several vaccines now in the development pipeline were not even contemplated in 1985. Rapid advances in the biomedical sciences and new knowledge about disease etiology and epidemiology can quickly change priorities. Before reviewing some of the major biomedical advances that are allowing vaccine development to proceed in ways not previously imagined, it will be useful to take a look at one of the vaccine successes known to all—the development of polio vaccines and the near eradication of a dreaded disease. Disease Burden Poliomyelitis was a relatively insignificant disease in the United States before 1900, when epidemics of increasing severity began to appear in different parts of the country. The average annual incidence of the disease for the years

OCR for page 17
Vaccines for the 21st Century: A Tool for Decisionmaking 1910 to 1914 was about 6 cases per 100,000 population. The frequency of paralytic polio continued to increase during the subsequent decades, and the rate for 1952 was approximately 37 cases per 100,000 population. In the early 1900s most deaths due to poliomyelitis were observed in infants under 1 year of age and in children 1 to 4 years of age. The death rate declined sharply in cases occurring at older ages (Health Information Foundation, 1959). However, by the early 1950s, poliomyelitis was observed with increasing frequency in school-age children and young adults. Around this period, sanitation and community conditions of hygiene began to improve significantly largely due to the efforts begun in the 1930s and 1940s. As living conditions improved, a good proportion of the population probably was not exposed to the virus until an older age. These observations may explain the apparent shift in the age of susceptibility to the development of paralytic disease between the 1900s and the 1950s (Health Information Foundation, 1959). Polio Vaccine Development Efforts to develop immunoprophylaxis against polioviruses began immediately after the isolation of the virus. Both killed and live virus candidate vaccines were developed as early as 1910, although at that time knowledge of the existence of three distinct poliovirus types was not available, and the fact that paralytic cases of polio represented only a tiny fraction of the total number of infections was not appreciated (Harrington, 1932; The National Foundation, 1961, 1962). For every known patient with paralysis, there may be as many as 100 to 1,000 patients who have subclinical infections (Harrington, 1932). During the early 1930s, studies were undertaken to vaccinate humans with infected monkey spinal cord suspensions inactivated with formalin or sodium ricinoleate (Brodie and Park, 1936; Kolmer et al., 1935). However, those trials failed because of a lack of adequate controls, the failure or inability to standardize vaccine preparations, and a lack of reproducible quantitative methods for virus titration. The battle against polio began seriously at the national level with the establishment of the National Foundation for Infantile Paralysis-March of Dimes organization in 1938. During World War II, information became available regarding the distinct antigenic types of the virus, their ability to induce specific antibody responses after inactivation, the ability of inactivated virus to induce protection against intracerebral challenge (Bodian, 1949; Morgan, 1948), and the capacity of polioviruses to replicate in vitro in human and primate tissue culture cells (Enders, 1952; Enders et al., 1949). Other wartime efforts directed toward the control of epidemics of influenza with an inactivated vaccine resulted by 1953 in a renewed interest in the development of formalin-inactivated poliovirus vaccines (Salk, 1953). The introduction of tissue culture techniques and the characterization of poliovirus passage in tissue culture were breakthroughs, and represent the cornerstone of current knowledge of cell-virus interaction. These observations significantly facili-

OCR for page 17
Vaccines for the 21st Century: A Tool for Decisionmaking tated the development of other live attenuated or inactivated candidate vaccines (Koprowski et al., 1952; Sabin, 1955). The inactivated type of poliovirus vaccine (IPV, also known as the “Salk vaccine”) was licensed in April 1955, and the oral, live-attenuated polio vaccine (OPV, also known as the “Sabin vaccine”) was licensed for human use in 1961–1962 (Commission on the Cost of Medical Care, 1964). With the introduction of polio vaccines, the incidence of poliomyelitis declined sharply. In 1965, only 59 cases of poliomyelitis were reported in the United States, and continued widespread use of OPV alone has essentially eliminated polio in the Americas. The last case of indigenously acquired wildtype poliovirus infection reported in the United States occurred in 1979 (Hinman et al., 1987). In 1985, the Pan-American Health Organization (PAHO) established the goal of eliminating poliomyelitis from the Western Hemisphere. The subsequent success of their efforts is best exemplified by the fact that the last confirmed case of paralytic poliomyelitis associated with wild-type virus infection occurred in Peru in 1991. The worldwide rate of routine immunization for polio increased from about 47% in 1985 to 80% in 1994. The annual number of cases of polio reported decreased from 39,361 in 1985 to about 6,241 in 1990, a decline of nearly 85%. Encouraged by these dramatic results, WHO established the goal of the global eradication of poliomyelitis by the year 2000 (CDC, 1995c). The worldwide eradication campaign is relying on use of OPV, but several countries in Europe successfully eradicated poliovirus with IPV only. Advantages of Evolving Vaccine Strategies As pointed out earlier, the large-scale use of IPV reduced the annual reported worldwide incidence of paralytic polio to about 0.8 cases per 100,000 population (about 900 cases) by 1961. At that time, more than 485 million doses of the vaccine had been distributed in the United States. Occasional cases of disease were reported in fully vaccinated subjects in the early phases of the vaccination program, so continued efforts in vaccine development led to improvements in its biologic activity and the introduction of enhanced-potency IPV (eIPV). Vaccine failures with eIPV are much reduced from the earlier version of the vaccine. No cases of paralytic disease have been reported in subjects successfully immunized with eIPV, which is extensively used as the vaccine of choice in many parts of the world. Since 1962, the United States had depended primarily on OPV in its polio eradication efforts. Between 1969 and 1983, however, about 225 cases of paralytic poliomyelitis associated with OPV were identified in the United States. An average of 8 to 10 cases of vaccine-associated paralytic polio continued to be reported in the United States each year. Despite the success and the benefits achieved with OPV and the continued absence of wild-type virusassociated disease, the rare occurrence of vaccine-associated paralytic disease

OCR for page 17
Vaccines for the 21st Century: A Tool for Decisionmaking has resulted in a change in societal and individual perspectives on the overall risk-to-benefit assessment for OPV in the United States (Plotkin, 1995). The availability of two complementary vaccines for the same disease allowed for a change in national recommendations in 1996: eIPV is now recommended for at least the first two of the four polio immunizations in the United States. This policy assures the protection of the population against natural infection of those immunized against the rare but real possibility of adverse effects of vaccination. As the polio success story demonstrates, active vaccine development efforts in the face of changing epidemiology, scientific advances in basic virology, licensure of more than one polio vaccine, improvements in existing polio vaccines, worldwide efforts at eradication, and re-evaluation of domestic vaccination policies have all been necessary to give the U.S. population nearly complete protection from the threat of polio virus at the close of the 20th century. ADVANCES IN BIOTECHNOLOGY AND MOLECULAR IMMUNOLOGY AND NEW OPPORTUNITIES FOR VACCINES As illustrated in the previous section, rapid scientific advances, changes in disease epidemiology, and development prioritization fueled the successful development of two different polio vaccines and the near eradication of a feared disease. Science has developed in an extraordinary fashion since the publication of 1985 IOM reports on vaccine development priorities (IOM, 1985a,b). These advances have occurred because of the development of molecular approaches to the cloning and characterization of virulence determinants of specific viral, bacterial, and parasitic organisms and because there is a better understanding of the cellular and molecular interactions that follow host responses to deliberate immunization or infection with a specific pathogen. This section presents a brief summary of some of the major scientific successes so that their contribution to the development of specific vaccines can be more fully appreciated. Fundamental Understanding of Helper T Cells for Antibody Versus CMI Responses, and for Cytotoxicity It may be useful to describe the development of regulatory T cells by simply considering mature T cells that are recent emigrants from the thymus and are naïve (e.g., they have not yet encountered an antigen as precursor T-helper cells). Note that precursors of Th cells normally recognize foreign peptides in association with major histocompatability complex (MHC) class II on antigenpresenting cells (APCs) and express the α:β T-cell receptor with a CD3+, CD4+, CD8− phenotype. On the other hand, precursors of cytotoxic T lymphocytes (pCTLs) express the α:β TCR. These pCTLs usually recognize foreign peptide in the context of MHC class I on target cells and normally exhibit a CD3+ CD4−,

OCR for page 17
Vaccines for the 21st Century: A Tool for Decisionmaking CD8+ phenotype. Thus, encounters with foreign antigens (peptides) result in the development of effector T cells that are either T-helper cell types for cellmediated immunity (CMI), delayed-type hypersensitivity (DTH), or antibody responses that lyse infected target cells (cytotoxic T lymphocytes [CTLs]). As T-helper cells mature in response to foreign antigens, they take on unique characteristics normally manifested by the production of distinct cytokine arrays. (Cytokines are nonantibody proteins released by a cell population, such as primal T lymphocytes, on contact with a specific antigen. They act as intercellular mediators, such as in the generation of an immune response.) Of great interest has been the finding that the environment and cytokine milieu greatly influence the further differentiation of T-helper cells. For example, stimulation by certain pathogens such as intracellular bacteria often leads to the formation of T-helper type (Th1) cells producing gamma interferon, interluken 2 (IL-2) and tumor necrosis factor beta, and these T cells often develop following the production of IL-12 by activated macrophages (Hsieh et al., 1993), presumably following ingestion of the particular intracellular pathogen or viral pathogen. In vivo, Thl-type immune responses are associated with the development of CMI and DTH responses and B-cell responses characterized by IgG2a antibody synthesis. On the other hand, exogenous antigen can also induce a unique CD4+ T-cell subset to produce IL-4 (Seder and Paul, 1994), which can trigger the formation of Th2-type cells that produce IL-4, IL-5, IL-6, IL-9, IL-10, and IL-13 (Coffman et al., 1991; Mosmann and Coffman, 1989; Seder and Paul, 1994). This latter array of cytokines is conducive to B-cell switches from secretary IgM (sIgM) expression to certain IgG subclasses and IgE. Furthermore, Th2 cells are considered to be the major helper cell phenotype for support of IgGl, IgG2b, IgE, and IgA responses in mice. CTLs have been shown to be important for eliminating virus-infected cells for clearing of infection (Taylor and Askonas, 1986; Yap et al., 1978; Zinkernagel and Doherty, 1979; Zweerink et al., 1977). It is generally accepted that the endogenous viral peptide processing that occurs during natural infection is a major pathway for the induction of effector CTLs. Most virus-specific CTLs are CD4−, CD8+ and recognize viralus peptides in association with MHC class I expressed on infected target cells. Since CTLs have been shown to be important effector cells for eliminating virus-infected cells, it will be of considerable importance to continue to determine the significance of antigen-specific CTL responses in mucosa-associated tissue, where most virus infections actually occur (see below). In this regard, several studies have shown that cell-mediated cytotoxicity, antibody-dependent cytotoxicity, and natural killer cell activity can be found in mucosa-associated tissues for immunity at sites of initial infection (Davies and Parrott, 1981; Ernst et al., 1985).

OCR for page 17
Vaccines for the 21st Century: A Tool for Decisionmaking Major Advances in Mucosal Immunity The mammalian mucosal immune system is an integrated network of tissues, lymphoid and constitutive cells, and effector molecules that protect the host from infection of the mucous membrane surfaces. This signifies a major difference from the peripheral immune system, where lymphoid cells and effector molecules are confined to individual lymph nodes, and intercommunication occurs by cell trafficking through the lymphatic and blood circulations. The induction of peripheral immune responses by parenteral vaccination does not result in significant mucosal immunity; however, the reverse is not true. Induction of mucosal immune responses can result in protective immunity in the peripheral immune compartment as well. The mucosal immune system is anatomically and functionally divided into sites where foreign antigens are encountered and selectively taken up for the initiation of an immune response and the more diffuse collection of B and T lymphocytes, differentiated plasma cells, macrophages, and other antigenpresenting cells, as well as mast cells that comprise effector tissues for mucosal immunity. This network is highly integrated and tightly regulated. The outcome of mucosal tissue encounters with foreign antigens and pathogens can range from mucosal and serum antibody responses and T-cell-mediated immunity to systemic anergy to oral or intranasal antigen, a response that is now termed mucosal tolerance. The separation between the mucosal immune system and the peripheral immune system has evolved as a major host defense mechanism. Mucosal surfaces are enormous, approximately 300 to 400 m2 and as such they require a significant expenditure of lymphoid cell elements for immunity. In this regard, the major antibody isotype in external secretions is IgA, and approximately 40 mg of IgA per kg of body weight is made in mucosal effector tissues each day, especially in the gastrointestinal tract (Conley and Delacroix, 1987). When this output of IgA is combined with its synthesis in bone marrow and peripheral lymphoid tissues, this isotype represents twice the amount of other isotypes combined, including the IgG subclasses, which are produced in higher mammals. Despite this propensity to produce IgA, the major effector cells in the mucosal immune system are T lymphocytes of both CD4+ and CD8+ phenotypes, and in some cases they can represent up to 80% of the entire cell population, clearly indicating their importance in mucosal immunity. The use of vaccines that induce protective mucosal immunity thus becomes attractive when one considers that most infectious agents come into contact with the host at mucosal surfaces. Induction of mucosal immune responses may not only protect the host from morbidity and mortality due to infection but may possibly prevent infection altogether. The childhood immunization schedule recommended by the Centers for Disease Control and Prevention (CDC, 1999) lists seven vaccines that children should receive: (1) HBV; (2) diphtheria and tetanus toxoids and pertussis (DPT); (3) Hib; (4) poliovirus; (5) measles, mumps, rubella; (6) varicella; and (7) rotavirus. Of those vaccines, OPV and rotavirus are administered by the mucosal route. In fact, of 30 classes of vaccines, toxoids,

OCR for page 17
Vaccines for the 21st Century: A Tool for Decisionmaking and proteins currently licensed for use in the United States, only 4 (OPV, rotavirus, adenovirus, and typhoid) are administered orally (CDC, 1994b) and none are given intranasally. Although parenterally administered vaccines induce protective immune responses, they rarely, if ever, induce mucosal immune responses that may prevent infection at the site of initial contact between the host and the infectious agent. The following sections describe some of the cellular and molecular components of the mucosal immune system of relevance to current mucosal vaccine strategies. Mucosal Immune System Organization Generally, foreign antigens and pathogens are encountered through ingestion or by inhalation, and the host has evolved in these regions organized lymphoid tissues that facilitate their uptake. These inductive sites contain B and T lymphocytes that in the presence of appropriate antigen-presenting cells, respond to the encountered antigen by developing into effector and memory B and T cells. These antigen-specific B- and T-cell populations then emigrate from the inductive environment via lymphatic drainage, circulate through the bloodstream, and home to mucosal effector regions using distinct homing receptors that recognize mucosal addressins. Thus, mucosal effector sites include these more diffuse tissues where antigen-specific T and B lymphocytes ultimately reside and perform their respective functions (i.e., cytokine and antibody synthesis, respectively) to protect mucosal surfaces. After the initial exposure to antigen in mucosal inductive sites, mucosal lymphocytes leave the inductive sites and home to mucosal effector tissues. Antigen-specific mucosal effector cells include IgA-producing plasma cells as well as B and T lymphocytes. Polymeric, usually dimeric, IgA is the primary immunoglobulin involved in the protection of mucosal surfaces and is locally produced in the gastrointestinal and upper respiratory tracts, nose, middle ear, gall bladder, uterine mucosa, and biliary tree as well as glandular tissues such as salivary, lactating mammary, prostate, and lacrimal glands (Phillips-Quagliata et al., 1994). The observation that antigen-specific S-IgA responses may be detected at mucosal surfaces other than the inductive site where antigen uptake initially occurred led to the discovery of the common mucosal immune system. Studies to elucidate the common mucosal immune system pathway showed that immunization of one mucosal inductive site could induce mucosal immune responses in all mucosal effector tissues. The common mucosal immune system provides a unique opportunity to develop mucosal vaccines that can be delivered orally or intranasally but that subsequently result in mucosal immunity at sites where immune protection is most desirable. One major hallmark of the mucosal immune response is the presence of IgA antibodies at mucosal surfaces. The importance of IgA transport across epithelial surfaces to external secretions should be considered when vaccines are being designed to prevent infections that occur at mucosal surfaces. Passive transfer

OCR for page 17
Vaccines for the 21st Century: A Tool for Decisionmaking studies with antigen-specific monoclonal IgA antibodies have provided evidence that antigen-specific IgA alone was able to protect against intranasal influenza virus infection (Renegar and Small, 1991), intestinal infection with Vibrio cholerae or S.typhi (Winner et al., 1991; Michetti et al., 1992), and gastric infection with Helicobacter felis (Czinn et al., 1993). Antigen-specific IgA presumably forms immune complexes with the colonizing pathogen and thereby inhibits the interaction of the pathogen with host epithelial cells, a protective mechanism known as immune exclusion (Mestecky and McGhee, 1987). In fact, passive transfer of monoclonal IgA antibodies by a backpack hybridoma method provided protection against mucosal challenge with virulent organisms but was generally unable to prevent infection when the organisms were introduced parenterally, suggesting that mechanisms for protection at a mucosal surface do not correlate with protection from systemic challenge (Michetti et al., 1992). Therefore, induction of antigen-specific IgA responses may provide a means of totally preventing bacterial infections or at least greatly reducing the size of the infectious inoculum at the sites of initial contact between most infectious agents and the host, the mucosal surfaces. Molecular Aspects of Virulence and Design of Recombinant Protein Vaccines The use of recombinant techniques for the production of protein-based vaccines as well as three-dimensional immunogenic structures is well exemplified by the safe and effective recombinant HBV vaccine. Even though an effective plasma-derived vaccine for HBV has been available for many years, the recombinant DNA-derived vaccine has resulted in two licensed vaccines, the baculovirus- and yeast-derived HBV vaccines. This example can be extended to virtually all current killed or partially purified bacterial or viral vaccines. Improvements in DPT involve the use of recombinant partial structures for diphtheria (CRM 197) and tetanus (fragment C) toxoids as well as partially purified proteins from Bordetella pertussis, the so-called acellular pertussis vaccine. The acellular pertussis vaccines were found to be effective following several trials in Italy, Sweden, and Germany during the past 3 years. This new generation rDaPrT vaccine should be followed by a completely recombinant form of pertussis vaccine in the next few years. Another important feature of new, recombinant vaccines involve the use of the insert baculovirus system to express genes for several proteins that comprise the virus capsid, which encloses the nucleic acid. The nucleic acid-free, viruslike particles (VLPs) represent important structures for vaccine development. For example, most pathogenic viruses express ligands in capsid proteins for receptors on host epithelial cells, which are the major initial site for virus entry into the host. In many instances, the VLPs also retain the receptor binding ligand that allows their uptake into a mucosal inductive site. As discussed above, the mu-

OCR for page 17
Vaccines for the 21st Century: A Tool for Decisionmaking cosal and systemic lymphoid cell systems are distinct and exhibit separate modes for the induction of immunity. The use of particulate VLPs that bind epithelial cells and M cells of mucosal sites such as the Peyer’s patches or GALTs may represent ideal modes for the induction of mucosal immunity to viruses. Although polysaccharide vaccines are usually not very immunogenic in infants, the titers of antibodies are increased by covalent coupling to protein carriers such as tetanus or diphtheria toxoid. Studies indicate that conjugates elicit T-cell-dependent antibody responses characterized by higher titers and switching to non-IgM isotypes. The current major conjugate vaccines have been developed to Hib or to several different capsular polysaccharide types of Streptococcus pneumoniae, and these represent important advances that have used molecular immunology, chemistry, and infectious disease expertise. Novel Vaccine Delivery Systems A number of major breakthroughs have occurred in the development of particulate vaccines. These range from lipid- or detergent-based enclosures; for example, liposomes (Gregoriadis, 1990) and immune-stimulating complexes (ISCOMS) (Morein et al., 1984; Claassen and Osterhaus, 1992), to chemical polymers, for example, microspheres (Eldridge et al., 1990; O’Hagan et al., 1993). Microphages are briefly described here to illustrate the promise of inert particles for vaccine delivery. Several types of microspheres have been used. The coating material is usually a biodegradable polymer, and methods for microencapsulation usually involve the separation of two-phase polymers. Emulsions from solvent evaporation-extraction are common. The microspheres produced in this way are spherical. Those ranging from 1 to 10 g are most effectively taken up by antigen-presenting cells as well as M cells in mucosal inductive sites (Eldridge et al., 1990). In general, the copolymer poly (dl-lactidecoglycolide) has been more extensively studied, and variations in polymer ratios can affect the rate of hydrolysis and antigen release. Other promising microspheres involve the polyphosphacenes-polyalignates. Because of their hydration properties, they are amenable to encapsulation of proteins in physiologic buffers, which avoid denaturation. Salmonella strains were the first to be used to investigate the potential for expression and delivery of recombinant vaccine proteins to the host immune system. This approach has been elegantly extended to other enteric gramnegative organisms, that is, recombinant V. cholerae and Shigella—as well as to recombinant bovine growth hormones, and to commensal bacteria—for example, oral streptococci and Lactobacillus, which is present in yogurt. However, recombinant Salmonella has remained the prototype for this approach and can be used to illustrate the principle (Hone et al., 1991; Chatfield et al., 1992; Tackett et al., 1992; Roberts et al., 1994). An important benefit of using recombinant Salmonella is that it remains an enteric bacterium that, following oral admini-

OCR for page 17
Vaccines for the 21st Century: A Tool for Decisionmaking stration, colonizes and penetrates the mucosal inductive tissues via the M cell and presumably delivers recombinant proteins to elicit mucosal immunity (see below). Current attenuation mutations involve deletions of two or more genes and are designed to avoid complementation by the host or other indigenous flora. A major advantage of this approach is that oral immunization with recombinant Salmonella can elicit protection from S.typhi, and at present several vaccines relying on recombinant Salmonella strains are in human phase I and phase II trials. Major breakthroughs have also been made in the development and use of former viral pathogens for the delivery of foreign antigens. Vaccinia virus has been used successfully for several antigens and cytokines. However, since most viruses impinge on the epithelium, which lines the mucosal immune system, more attention has been paid to the development of mucosal virus delivery. Two examples will suffice: recombinant poliovirus, since this has been a successful oral vaccine; and recombinant adenovirus (also a successful respiratory pathogen) for delivery to the mucosal tissues of the gastrointestinal tract as well as to the upper respiratory tract and lungs. The success and efficacy of OPV make it an attractive vector for the delivery of mucosal vaccines, particularly when immunity to enteric pathogens is desired. This vaccine induces both mucosal and systemic immune responses and offers protection from infection. Polio-virus-specific MHC class II-restricted CD4+ T cells in peripheral blood mononuclear cells from orally vaccinated individuals have also been detected. This finding suggests that poliovirus can be used as an antigen delivery vehicle to induce CD4+ T-helper cells that can regulate mucosal IgA B-cell responses, in addition to the typical virus-induced CTL type of immunity. Recently, investigators have used chimeric poliovirus genomes in which gag and pol genes of human immunodeficiency virus type 1 were substituted for the VP2 and VP3 outer capsid genes of poliovirus (Porter et al., 1993; 1995). Transfection of the minireplicon genomes containing the gag or pol gene into cells produced the appropriate fusion protein. These minireplicons were then encapsulated and amplified by transfecting them into cells previously infected with a recombinant vaccinia virus that expresses poliovirus capsid precursor protein P1. For immunization studies, the encapsulated replicons were passaged in the presence of poliovirus type 2 Lansing, which again resulted in encapsidation of the replicons by the capsid proteins provided by poliovirus. These replicons were then given to mice by the intramuscular, intrarectal, or oral routes, and the mice were later boosted by the same route. Results from these studies indicated an increased production of antipoliovirus antibodies in serum and increased virus-specific IgA antibodies in saliva and in gastrointestinal tract secretions. In addition, the detection of anti-Gag and anti-Env antibodies in serum after intramuscular immunization and in external secretions (Moldoveanu et al., 1995) following mucosal delivery has clearly established the immunogenicities of the minireplicons.

OCR for page 17
Vaccines for the 21st Century: A Tool for Decisionmaking Delivery of vaccines by intranasal immunization has shown that this is a very effective route for the induction of respiratory and parenteral immune responses. However, few recombinant, avirulent viruses and no bacteria are available to take advantage of this novel mode of immunization. One major exception is the adenovirus vectors, which have generated interest not only as a means for mucosal immunization but also as potential vectors to transfer the corrective CFTR gene for the treatment of cystic fibrosis. The major advantages of adenovirus are its cloning efficiency and its ability to accommodate large foreign DNA sequences. In this regard, most adenovirus-based vectors have been derived from group C adenoviruses, and the first generation of adenovirus vectors were made replication-defective by deletion of the viral one region (Yang et al., 1994). This region encodes the immediate-early gene products and is required for the initiation of viral replication. Although adenovirus vectors were rendered replication defective, a number of studies have shown that these vectors can induce unwanted inflammatory responses (Yang et al., 1994). Second-generation vaccines with additional deletions should obviate some side effects. It should be noted that an adenovirus vaccine has been given by the oral route to military recruits and has been successful. The next challenge will be to use the attenuated adenoviruses as vectors for effective intranasal immunization. Recent studies have shown the feasibility of using recombinant plants for the generation of vaccines. Initial work with the tobacco plant showed that foreign gene expression could be accomplished. However, of more importance, it has now been shown that potato tubers may be used to express proteins including Escherichia coli labile toxin B subunit, rotavirus VLPs and HBV antigen (Haq et al., 1995; Thanavala et al., 1995). At present, the level of expression of recombinant protein is relatively low; however, it is anticipated that much higher levels can be achieved, making this approach one of the most promising ways of devising and producing an oral vaccine. Another interesting benefit from transgenic plant technology has been the production of a functional IgA antibody molecule (Ma et al., 1995), which again offers an alternative approach to the production of vaccines for passive mucosal immunity. It has been known for a long time that the use of live, attenuated vaccines results in more appropriate and protective immune responses than does the use of inactivated vaccines. Expression of antigens in the host results in the correct protein conformation and glycosylation patterns. Even more important, intracellular protein processing can allow presentation by the class I MHC for effective CTL responses. A limitation to the development of vaccines against viruses such as the influenza virus is the diversity of viral envelope proteins among different strains. Therefore, efforts in vaccine development have focused on induction of memory CTLs that react to epitopes shared by different strains of virus. Most efforts to generate CTL responses have used replicating vectors to either produce the antigen in the host cell or to deliver peptides into the cytoplasm. However, the selection of peptide epitopes presented by MHC molecules is dependent on the structure of individual MHC molecules, and the peptide approach has been shown to have some limitations in humans.

OCR for page 17
Vaccines for the 21st Century: A Tool for Decisionmaking As an alternate method of immunization against influenza virus (e.g., use of a plasmid), DNA vaccines have become promising approaches for the protection of mucosal surfaces. Like recombinant vectors, the transfected DNA results in the presentation of antigenic epitopes in association with the class I MHC. In addition, the significant advantages of using gene transfer technology for mucosal immunization against a pathogen such as the influenza virus are that (1) no infectious agents are being used, (2) combined vaccines are easily and rapidly made, and (3) DNA stability is not affected by high temperatures and therefore is more suitable as a vaccine in less developed countries. The feasibility of polynucleotide vaccines was first shown in studies in which plasmid DNA was directly injected into the quadriceps of mice. Many recent studies have shown that protection against mucosal pathogens may be achieved by DNA immunization. Most DNA immunization protocols performed so far have used inoculation of the DNA into muscle cells or particle bombardment into dermal or epidermal cells. However, in nature, most foreign antigens are first confronted by the mucosa. Thus, gene administration to the mucosal surfaces would mimic exposure to most pathogens and may more efficiently induce a protective immune response. In this regard, it has been shown in mice that intranasal inoculation of a plasmid expression system for influenza virus hemagglutinin induces resistance to lethal challenge with live influenza viruses. This technology has more recently been extended to other viruses including antigen components of human immunodeficiency virus, simian immunodeficiency virus, and rabies virus, among others. Thus far, most preclinical results have been promising, and phase I trials with DNA vaccines are now under way. Recent Advances in Development of Novel Adjuvants Most vaccine antigens yield only weak immune responses when given by themselves either parenterally or orally. Thus, the generation of an effective immune response usually requires the addition of an adjuvant, which is a substance that enhances the immune response. Adjuvants have been shown to affect virtually every measurable aspect of antibody responses, including the kinetics, duration, quantity, isotype, avidity, and generation of neutralizing or protective antibodies. Certain adjuvants can enhance T-cell-mediated immunity, including both delayed-type hypersensitivity mediated by CD4+ cells and CTL responses mediated by CD8+ cells. However, fewer adjuvants tend to stimulate cell-mediated immune responses than to stimulate antibody formation. Although adjuvants have been used empirically for many years, the mechanisms by which they act are not well understood, partly because the adjuvants themselves have been very complex, making such evaluations difficult (Waksman, 1979). The best-understood adjuvants have a multiplicity of effects

OCR for page 17
Vaccines for the 21st Century: A Tool for Decisionmaking on immune cells, and different adjutants have very divergent effects on the same cells. Waksman (1979) made the point that one must define the target cell affected by the adjuvant and the cellular and molecular modes of action of the adjuvant on the target cell. This information is either rudimentary or nonexistent for most adjuvants. Recently, more highly purified molecules have been isolated from traditional adjuvants, such as muramyl dipeptide from mycobacteria and monophosphoryl lipid A from endotoxin. This may simplify the dissection of their effects. Although many if not all of these mechanisms are likely to apply to mucosal adjuvants as well, there is little information on agents with mucosal adjuvanticity, despite a great need for such agents. Mucosal immunization is the route of choice for protection from many pathogens, but the development of effective mucosal vaccines has lagged, in part due to a lack of suitable mucosal adjuvants. Most protein antigens are not only poor immunogens when given mucosally but also induce tolerance instead of immunity. Mucosal adjuvants are needed to overcome this potential outcome of mucosal antigen exposure. Cholera toxin has been shown to enhance the immunogenicity of relatively poor mucosal immunogens when it is mixed or conjugated together and given orally; thus, cholera toxin and its B subunit have generated a great deal of interest as potential adjuvants for vaccines. Scientific Rationale for Vaccines against Autoimmune Diseases During this past century, the number of diseases attributable to the body’s own self-reactivity has risen, so that now it is recognized that there are autoimmune diseases directed against every organ, as well as systemic diseases affecting a broad range of systems. Examples of these diseases include diabetes, rheumatoid arthritis, multiple sclerosis, thyroiditis, myocarditis, and systemic lupus erythematosus. Within the last 15 years, the subfield of autoimmunotherapy within immunology has made impressive strides, along with the detailed knowledge of the initiation and propagation of autoimmune diseases. Nevertheless, almost to the present, the major treatments have been (1) generic anti-inflammatory agents such as steroids, which have well-known, serious side effects; or (2) cytostatic and cytotoxic drugs whose nonspecific effects on unrelated systems lead to additional morbidity and mortality. In model animal systems, a wide variety of specific therapies in combination with the above general agents have been explored successfully with the aim of transferring this technology to human use. Several general approaches have been attempted in the effort to either prevent or to alter the course of autoimmune diseases. One strategy is to employ specific antigens or peptides for the induction of immune tolerance among the relevant cells. This approach requires that the antigen or peptide can be defined, although more recently it has been clear that bystander regulation induced by a specific agent could be successful in preventing initiation of other immunospeci-

OCR for page 17
Vaccines for the 21st Century: A Tool for Decisionmaking fic clones. A second strategy is through affecting T-cell subset choice by deviation of the T-cell system from one differentiation arm to another, such as Thl to Th2 or Th3. The change in cytokine pattern produced by the same clone might be able to prevent specific damage, as well as provide protection by bystander suppression. A third strategy is the use of regulatory peptides derived from the T-cell receptors themselves, which are employed by the organism to bring about homeostasis through suppression of unwanted reactivity. Fourth, treatments directed against an important receptor or its mediator can have a curative effect in some cases. Finally, viral vectors carrying genes coding for specific antigen products, and/or ameliorating agents such as cytokines and chemokines have been shown to be potent response modifiers. Each of these classes of agents will now be considered in greater detail. Immune Tolerance Induction If immune tolerance is defined as the state in which response to a particular immunogen is absent, this would include cases in which the specific precursor T or B cell is deleted/anergized, or exhausted through chronic expansion of precursors. It would also include immune deviation where the cell in question does not become immunologically silent but rather follows a different functional program. In fact, systemic introduction of a native antigen or peptide, or one that has been modified to increase its affinity for the MHC or TcR, can induce deletion of clones with the highest affinity for the antigen. A feature of self-antigens is that under usual conditions, the B and especially the T cells directed against the most dominant self determinants will have been removed owing to deletional mechanisms in the thymus and periphery, leaving a rather large assemblage of lymphocytes directed against subdominant and cryptic determinants. This protected repertoire can be engaged under highly inflammatory conditions. The secondary determinants, which appear to then be the dominant ones, and which can activate this protected repertoire, can be removed by appropriate antigenic administration, and this may suffice to prevent autoimmunity. Another mechanism of T-cell deletion is brought about by apoptosis of specific lymphocytes, via antigen-induced T-cell death. Apoptosis can represent the final mechanism of both exhaustion and deletion. One important feature of autoimmune pathology is that the initial response, which may be directed against a single or small number of antigenic determinants, then diversifies to include many more determinants and a broader repertoire of T cells, in what has been termed intramolecular and intermolecular determinant spreading. Thus, it appears to be important to regulate T cells directed against the initiating determinant. It has also been shown that tolerance induction to the secondary determinants plays a role in the prevention of spreading. It becomes necessary to consider how the response that has been established can be diverted effectively, and this will be the most difficult aspect of therapeutic

OCR for page 17
Vaccines for the 21st Century: A Tool for Decisionmaking vaccines. So far, even in model systems, the best results have been obtained by treatments close to the beginning of the appearance of symptoms. Oral and nasal tolerance induction are very effective in regulating autoimmune diseases in animal models, and show some promise in human trials. Administration of antigen by these routes leads to a combination of effects, including deletional and anergic consequences as well as immune deviation. A deviated response that is often curative in animal models of diabetes and multiple sclerosis, for example, involves a switch from a Thl to a Th2 direction, and will be discussed below. Much of the benefit from the tolerance induced through the mucosal route may occur by deviation. It can be hoped that for clinical purposes, when the inciting antigen(s) is (are) known, tolerance induction may lead to a broadened effect via bystander suppression or regulatory spreading. Immune Deviation It is evident from model systems that the induction of a Th2 (anti-inflammatory) or a Th3 (regulatory cytokine) state of differentiation can prevent or down-regulate an autoimmune disease course. One of the favored ways of accomplishing this is through mucosal introduction of antigen, which generally deviates responses in a Th2 or Th3 direction. In the type I diabetes of the non-obese diabetic (NOD) mouse, treatment involving deviation has been shown to prevent the disease at a time well after its initiation, in the midst of increasing insulitis. This occurs in what is conceived to be a Thl mouse strain, particularly disposed towards inflammatory induction; in other strains of mice of the Th2 type, such as the BALB/c, certain autoimmune diseases are difficult to induce. The nature of the peptide chosen for therapy is an important ingredient: for example, determinants with high affinity for the MHC tend to induce Thl responses. On the other hand, altered peptide ligands may be designed so as to induce a Th2 or regulatory response. In allergic individuals, the reverse deviation, from Th2 to Thl, may be an effective route to therapy. Receptor-Centered Regulation The intrinsic regulatory properties of T- and B-cell circuits can be employed in vaccines. Accordingly, it has been shown in lupus that T cells that modulate B-cell activation can be induced with peptides derived from the B-cell receptor. Likewise, in such diseases as multiple sclerosis and its animal equivalent, experimental autoimmune encephalomyelitis (EAE), or collagen arthritis, regulatory T cells specific for antigenic determinants on the aggressive T-cells receptor have been shown to exert a curative effect. In the B10.PL mouse strain, which demonstrates a single spike transient EAE, such TcR-specific regulatory T cells can be demonstrated when disease disappears, and not earlier. Recently it

OCR for page 17
Vaccines for the 21st Century: A Tool for Decisionmaking has been shown that the consequence of these regulatory circuits is to divert the effector population from the dangerous Thl state to a protective Th2 direction. Targeting Cytokines or Their Receptors Antibodies directed against cytokines or their receptors may appear to be just one step removed from the generic type of steroid therapy. Nevertheless, applied at the right time, an agent such as anti-TNF (tumor necrosis factor) antiserum can exert a remarkable effect on patients with rheumatoid arthritis. The success of this treatment has led to more efficacious approaches now in clinical trials that counteract the effects of TNF, such as ligands that bind preferentially to the TNFa receptor. Viral Vectors in Autoimmune Therapy Gene therapy using viral vectors can be used to modulate autoimmune disease, not in an effort to permanently alter the recipient, but rather to provide a localized, potent, short-acting agent—more like a molecular medicine. For example, treatment with an adenovirus bearing IL-12 genes can serve as a protective vaccine for a Th2 mouse, which is unable to raise a strong, protective Thl response against a microorganism. Such animal experiments suggest that in treatments of the near future, combinations of agents provided through genetic alteration of a viral vector will be used as prime modifiers of disease. Preventive Vaccines for Autoimmune Diseases Preventive vaccines for autoimmune diseases are also on the near horizon. Individuals who are genetically susceptible to type I diabetes can be identified now, and quite early in the disease course, when diagnostic antibodies appear to antigens of the islets of Langerhans, tolerance-inducing therapy to insulin or glutamic acid decarboxylase, two major candidate diabetogens, can be introduced. In other diseases in which the inciting antigens are known, immune deviation may be started at the very first signs of disease.