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 32
Immunization Safety Review: Multiple Immunizations and Immune Dysfunction committee and cited in this report are available—in the form reviewed by the committee—through the public access files of the National Academies at 202– 334–3543 or www.national-academies.org/publicaccess.) UNDER REVIEW: MULTIPLE IMMUNIZATIONS AND IMMUNE DYSFUNCTION Over the past two decades, the pediatric immunization schedule has grown more complicated. In 1980, the youngest infants received vaccines against four diseases (diphtheria, tetanus, pertussis, and polio). Today, a healthy child immunized in complete accord with the recommended childhood immunization schedule receives up to 15 doses of five vaccines to protect against seven diseases by 6 months of age and up to 20 doses of seven vaccines to protect against 11 diseases by 2 years of age (see Figure 2). Furthermore, the immunization schedule seems likely to expand in the next decade, with more vaccines for infants and children being developed or considered. The increase in the number of vaccines and vaccine doses given to children has led to concerns among some about possible adverse effects of individual vaccines or of the aggregate vaccine exposure. One such concern has been prompted by increased incidence of conditions associated with immune system dysfunctions—for example, asthma and type 1 diabetes, often referred to as insulin-dependent diabetes mellitus. Although genetic factors are known to affect the risk of these diseases, increases in their incidence seem more likely to reflect changes in environmental exposures than in the genetic makeup of a population. Increased exposure to vaccines has been proposed as one possible environmental modifier of immune function. For others, however, the concern is that having to administer many injections in a short period of time could adversely affect the acceptance of the immunization schedule by parents and health care providers, leading to reduced vaccination rates and greater risk of vaccine-preventable disease. Combination vaccine products can mean fewer injections per visit, but, they give little comfort to those who worry about the safety of multiple vaccine exposures. In addition, if adverse effects occur after the receipt of a combination vaccine product it may be difficult to determine which individual vaccine is responsible. Another concern is that development of novel vaccine delivery systems (i.e., nasal sprays and patches) may further complicate the issue of the effects of vaccines on the developing infant immune system. Framing the Question The Interagency Vaccine Group asked the Immunization Safety Review Committee to address the concern that multiple immunizations can adversely affect
OCR for page 33
Immunization Safety Review: Multiple Immunizations and Immune Dysfunction the developing immune system. To conduct its review, the committee had to establish a clear statement of the question before it, as well as a manageable scope of inquiry. Both “multiple immunization” and “immune system dysfunction” must be defined for the purposes of this report. First, multiple immunization has several possible meanings. A single dose of vaccine may present multiple antigens for a single disease (e.g., polio or pneumococcal vaccines) or multiple antigens for multiple diseases (e.g., measles-mumps-rubella [MMR] vaccine). Also, individual doses of several separate vaccines may be administered at a single health care visit. And further “repeat” doses of a single vaccine are administered, alone or with other vaccines, at specified intervals (e.g., 2, 4, and 6 months of age). The committee intended its primary focus to be on exposure to multiple vaccine antigens during infancy and childhood. However, as described in the section below on causality, the literature base is not large, and relevant studies often addressed the effects of incremental exposure differences, such as four vaccines compared to three. The committee restricted its considerations to those vaccines used in the United States. Thus, data regarding BCG vaccine, which is used against tuberculosis in other countries, did not contribute directly to the committee’s causality arguments. (Studies of BCG did, however, help inform the committee’s understanding of the biological arguments for and against the hypotheses.) Nor did the committee address possible effects of smallpox vaccine, which has not been used in the United States for 30 years. The committee included studies of “one vaccine” if it contained antigens against more than one disease or more than one strain of infectious agent. For example, the diphtheria and tetanus toxoids and pertussis (DTP) vaccine—whether whole-cell (DTwP) or acellular (DTaP) preparations—would be considered to represent a “multiple immunization,” as would the polio vaccines, which contain live or killed viruses against three distinct strains of poliovirus. Second, immune system dysfunction is a broad term. A brief review of the literature about immunization safety indicates that three types of immune system injury are of concern to vaccine safety advocates: risk of infection, risk of allergic diseases, and risk of autoimmune diseases. These concerns have gained prominence due to a generic consideration of biological mechanisms and due to studies, mostly ecological analyses, that are occasionally salient in the lay and scientific literature. The committee considered two possible pathways to adverse outcomes: stimulation of harmful immune responses or suppression of beneficial immune responses. The committee addressed infections only as distinct from those the vaccines are intended to protect against—referred to as heterologous infection—and in lieu of trying to sweep broad categories of allergic and autoimmune diseases, the committee narrowed its focus to specific conditions. It appeared to the committee that much of the concern, and a large component of the evidentiary base, centered around the allergic disease of asthma and the autoimmune form of diabetes—that is, type 1a diabetes, one of two types of
OCR for page 34
Immunization Safety Review: Multiple Immunizations and Immune Dysfunction FIGURE 2 Recommended Childhood Immunization Schedule, United States, 2002
OCR for page 35
Immunization Safety Review: Multiple Immunizations and Immune Dysfunction NOTES 1. Hepatitis B vaccine (Hep B). All infants should receive the first dose of hepatitis B vaccine soon after birth and before hospital discharge; the first dose may also be given by age 2 months if the infant’s mother is HBsAg-negative. Only monovalent hepatitis B vaccine can be used for the birth dose. Monovalent or combination vaccine containing Hep B may be used to complete the series; four doses of vaccine may be administered if combination vaccine is used. The second dose should be given at least 4 weeks after the first dose, except for Hib-containing vaccine which cannot be administered before age 6 weeks. The third dose should be given at least 16 weeks after the first dose and at least 8 weeks after the second dose. The last dose in the vaccination series (third or fourth dose) should not be administered before age 6 months. Infants born to HBsAg-positive mothers should receive hepatitis B vaccine and 0.5 mL hepatitis B immune globulin (HBIG) within 12 hours of birth at separate sites. The second dose is recommended at age 1-2 months and the vaccination series should be completed (third or fourth dose) at age 6 months. Infants born to mothers whose HBsAg status is unknown should receive the first dose of the hepatitis B vaccine series within 12 hours of birth. Maternal blood should be drawn at the time of delivery to determine the mother’s HBsAg status; if the HBsAg test is positive, the infant should receive HBIG as soon as possible (no later than age 1 week). 2. Diphtheria and tetanus toxoids and acellular pertussis vaccine (DTaP). The fourth dose of DTaP may be administered as early as age 12 months, provided 6 months have elapsed since the third dose and the child is unlikely to remm at age 15–18 months. Tetanus and diphtheria toxoids (Td) is recommended at age 11–12 years if at least 5 years have elapsed since the last dose of tetanus and diphtheria toxoid-containing vaccine. Subsequent routine Td boosters are recommended every 10 years. 3. Haemophilus influenzae type b (Hib) conjugate vaccine. Three Hib conjugate vaccines are licensed for infant use. If PRP-OMP (PedvaxHIB® or ComVax® [Merck]) is administered at ages 2 and 4 months, a dose at age 6 months is not required. DTaP/Hib combination products should not be used for primary immunization in infants at age 2, 4 or 6 months, but can be used as boosters following any Hib vaccine. 4. Inactivated poliovirus vaccine (IPV). An all-IPV schedule is recommended for routine childhood poliovirus vaccination in the United States. All children should receive four doses of IPV at age 2 months, 4 months, 6–18 months, and 4–6 years. 5. Measles, mumps, and rubella vaccine (MMR). The second dose of MMR is recommended routinely at age 4–6 years but may be administered during any visit, provided at least 4 weeks have elapsed since the ftrst dose and that both doses are administered beginning at or after age 12 months. Those who have not previously received the second dose should complete the schedule by the visit at age 11–12 years. 6. Varicella vaccine. Varicella vaccine is recommended at any visit at or after age 12 months for susceptible children (i.e. those who lack a reliable history of chickenpox). Susceptible persons aged =13 years should receive two doses, given at least 4 weeks apart. 7. Pneumococcal vaccine. The heptavalent pneumococcal conjugate vaccine (PCV) is recommended for all children aged 2–23 months and for certain children aged 24–59 months. Pneumococcal polysaccharide vaccine (PPV) is recommended in addition to PCV for certain high-risk groups. (See CDC, 2000b). 8. Hepatitis A vaccine. Hepatitis A vaccine is recommended for use in selected states and regions, and for certain high-risk groups; consult your local public health authority. (See CDC, 1999c). 9. Influenza vaccine. Influenza vaccine is recommended annually for children age =6 months with certain risk factors (including but not limited to asthma, cardiac disease, sickle cell disease, HIV and diabetes; (see CDC, 2001d) and can be administered to all others wishing to obtain immunity. Children aged =12 years should receive vaccine in a dosage appropriate for their age (0.25 mL if age 6–35 months or 0.5 mL if aged =3 years). Children aged =8 years who are receiving influenza vaccine for the first time should receive two doses separated by at least 4 weeks. Additional information about vaccines, vaccine supply, and contraindications for immunization, is available at www.cdc.gov/nip or at the National Immunization Hotline, 800–232–2522 (English) or 800–232–0233 (Spanish).
OCR for page 36
Immunization Safety Review: Multiple Immunizations and Immune Dysfunction what are often referred to as insulin-dependent diabetes mellitus (IDDM). The committee also considered neurological disorders for which the injury is known to be caused by the immune response, including MS and Guillain-Barré syndrome, but did not include these in its causality considerations due to the paucity of epidemiological/clinical information that addresses the possible role of multiple immunizations rather than individual vaccines. In addition, the committee will likely address at least some of these adverse outcomes in subsequent reports. The scope of the committee’s inquiry can be summarized in the following three questions: Do multiple immunizations have adverse short-term effects on the infant immune system that are reflected in increased susceptibility to heterologous infection? Does exposure to multiple antigens, as administered in vaccines, directly and permanently redirect or skew the immune system toward autoimmunity, as reflected in type 1 diabetes? Does exposure to multiple antigens, as administered in vaccines, directly and permanently redirect or skew the immune system toward allergy, as reflected in asthma? The committee was unable to address the concern of some that repeated exposure of a susceptible or fragile child to multiple vaccines over the developmental period may also produce atypical or nonspecific immune or nervous system injury that could lead to severe disability or death (Fisher, 2001b). Such adverse health outcomes may not be “classical” diseases but variants of diseases. Variants would not necessarily be picked up in epidemiological or clinical investigations that use strict diagnostic criteria. There are no epidemiological studies that address this, either in terms of exposure or outcome. That is, there is no study that compares an unvaccinated control group with children exposed to the complete immunization schedule, nor are there any studies that looked at health outcomes other than those classically defined, such as infections, allergy, or diabetes. Thus the committee recognizes with some discomfort that this report addresses only part of the overall set of concerns of some who are most wary about the safety of childhood vaccines. Key Features of Immune Response The immune system of humans and other vertebrates has the capacity both for generalized and specialized responses to organisms, such as bacteria, viruses, and parasites. Generalized responses are produced by mechanisms of innate immunity, while the mechanisms of adaptive immunity generate highly specialized responses to a diverse array of antigens; these are presented by microbes or by products such as vaccines, which may incorporate only specially selected
OCR for page 37
Immunization Safety Review: Multiple Immunizations and Immune Dysfunction antigens. The capacity for highly specialized immune responses carries with it the possibility that those responses will be directed against antigens of the body’s own cells, the process known as autoimmunity, or against normally harmless environmental materials, such as foods and pollens, a process known as allergy. Antigen-specific immunity is mediated by T and B lymphocytes (also referred to as T and B cells)1 and their products. These cells carry antigen-specific receptors on their surface. B lymphocyte receptors, the immunoglobulins (e.g., IgA, IgE, IgG), can potentially react with a wide variety of molecular structures. T lymphocyte receptors recognize short pieces of proteins (peptides) bound to self-major histocompatibility (MHC) molecules, which in humans are referred to as HLA (human leukocyte antigen) molecules. As an individual’s T and B cells develop, an enormous diversity of receptors is formed, allowing those cells to recognize and respond to the variety of antigens that might be encountered over a lifetime. This diversity is achieved by a nearly random process of genetic recombination of the genes that encode T- and B-cell receptors. When T and B cells are activated by antigens encountered through infection or vaccines, they multiply and differentiate into effector cells tailored to respond to those antigens. The effector T cells include two types of “helper” cells, designated Th1 and Th2. During their development and at all stages of their subsequent existence, T and B lymphocytes are “educated” by their environment. Initially, their antigen receptors have no intrinsic bias—that is, they are as likely to recognize antigens from the individual (self antigens) as from a foreign source (e.g., a microbe). As part of the education of T lymphocyte precursors in the thymus, those cells must show that their receptors are capable of reacting with self-peptides bound to self-MHC molecules. Cells that do not react or react very strongly to these self antigens usually die. Cells that react weakly are allowed to mature, leave the thymus, and go to the secondary lymphoid tissues as naïve T cells (naïve refers to the immune system at birth). This process ensures that T cells have the potential to be useful in that they can react with self-MHC. Similarly, strongly self-reactive B lymphocytes are removed during development. The censoring of more strongly self-reactive T and B lymphocytes is imperfect, however. As a result, some strongly self-reactive lymphocytes with the potential to produce autoimmunity can be found in the blood and secondary lymphoid tissues of most apparently normal individuals. Usually, these self-reactive T and B lymphocytes do not induce autoimmunity. For the most part, they simply do not encounter self antigens in a context that can trigger lymphocyte activation and expansion. Others are unable to respond even on encounter with self-antigen because they are anergic and/or short-lived cells that will soon die. Even when naïve self-reactive T cells do encounter self antigens, these 1 The designations of these cells reflect the sites where they mature. T lymphocytes migrate from the bone marrow to mature in the thymus, whereas B lymphocytes mature in the bone marrow.
OCR for page 38
Immunization Safety Review: Multiple Immunizations and Immune Dysfunction T cells are held in check in most individuals by counter-regulatory mechanisms, including suppressive cytokines and regulatory (suppressor) T lymphocytes that prevent the self-reactive T (and B) lymphocytes from responding (Ermann and Fathman, 2001; Letterio and Roberts, 1998; Maloy and Powrie, 2001; Moore et al., 2001; Roncarolo and Levings, 2000; Rook et al., 2000; Shevach, 2000; Wills-Karp et al., 2001; Zhang et al., 2001). Self-reactive B lymphocytes are also held in check by the lack of T cell help, without which they are unable to replicate and produce higher affinity antibodies in greater quantities. In some individuals, however, these regulatory processes fail, allowing self-reactive T and B cells to replicate, differentiate into effector cells, and cause autoimmunity—which is likely to be a multistep process. A common initial event may be the proliferation and differentiation of naïve self-reactive T (and in some cases B) cells into effector/memory cells. Thereafter, other mechanisms may further amplify the T cell response sufficiently to produce, sustain, or trigger a relapse of clinical autoimmune disease. There may also be cases where mixed Th1 and Th2 T cell responses can mediate autoimmunity (Benoist and Mathis, 2001; Marrack et al., 2001). Genetic factors have been shown, through a substantial body of data from both human studies and animal models, to play a critical role in determining risk for autoimmunity. In some rare disorders, single-gene defects are uniformly associated with the development of autoimmunity.2 However, the vast majority of human autoimmune diseases appear to be complex traits in which multiple genetic factors determine disease susceptibility and environmental factors determine whether disease develops (Ermann and Fathmann, 2001; Robles and Eisenbarth, 2001; Wanstrat and Wakeland, 2001). Familiar examples of such diseases include type 1a diabetes mellitus, systemic lupus erythematosus (SLE), and multiple sclerosis (MS) (Noseworthy et al., 2000; Robles and Eisenbarth, 2001; Steinman, 2001; Wakeland et al., 2001; Wanstrat and Wakeland, 2001; Wucherpfennig and Eisenbarth, 2001). Considerable progress has been made in identifying genetic factors that determine risk for autoimmune disorders. One major risk factor—common to type 1a diabetes, SLE, MS, and many other autoimmune diseases—is related to the MHC/HLA locus, which encodes the molecules that bind and present antigenic peptides to T lymphocytes. Genetic differences among individuals—in the ability of their MHC molecules to bind specific antigenic peptides in such a way that a portion of the peptide (the epitope3) is recognized by their T lymphocytes—influences the differences among individuals in the generation of T 2 Examples include two rare forms of polyendocrine autoimmunity (Aaltonen and Bjorses, 1999; Bennett et al., 2001; Wildin et al., 2001) and the autoimmune lymphoproliferative syndrome (Jackson and Puck, 1999). Similarly, more than 90 percent of individuals with a genetic absence of the complement protein Clq will develop systemic lupus erythematosus (Wanstrat and Wakeland, 2001). 3 Epitope is a “portion of an antigenic molecule that is bound by an antibody or gives rise to the MHC-binding peptide that is recognized by a T-cell receptor” (Parham, 2000).
OCR for page 39
Immunization Safety Review: Multiple Immunizations and Immune Dysfunction lymphocytes able to respond to self or foreign antigenic peptides. The other genes that confer risk for autoimmune disease are less completely characterized, but the best candidates are genes that regulate the amplitude and quality of the immune response or that affect the generation of specific antigenic epitopes. Atopy or allergy refers to diseases resulting from IgE-associated immune responses to innocuous environmental substances, such as certain foods or pollens. Allergic individuals have a hereditary predisposition to mount IgE responses when they encounter such substances, which are referred to as allergens, and they develop “atopic” (allergic) diseases such as asthma. This predisposition results in part from a bias favoring the generation of Th2 T cell responses to allergens, which produce the cytokines IL-4, IL-5, and IL-13 that favor the production of IgG4 and IgE antibodies by B cells and are implicated in allergic types of inflammation. By contrast, non-allergic individuals either do not mount an immune response to environmental allergens or mount a Th1 T cell response, which cells produce interferon-γ (IFN-γ) and favors the production of IgG1 antibodies. Multiple genetic factors appear to involved in the predisposition to atopy, but these are at present incompletely understood. In humans, the immune system begins development in early gestation. Although the human fetus has the potential to respond to foreign antigens by mid-gestation, exposure is very limited and the immune system is often referred to as naïve at birth. Active immunity in the neonate includes B and T cell responses, although the responses are not identical to those of older children. Infants’ B cell responses to T cell-independent antigens, such as polysaccharide antigens, is less vigorous than in adults. Thus, pure polysaccharides (including unconjugated H. influenzae and S. pneumoniae polysaccharides) do not induce an effective antibody response in children under approximately 2 years of age. However, if these polysaccharides are conjugated (linked) to protein antigens, they become T cell-dependent and induce protective antibody responses even in young infants. The effectiveness of such vaccines reflects the substantial maturity of T cell and T cell-dependent B cell responses, and the diverse repertoire of antigens that can be recognized by T cell and B cell receptors in the human infant. There are certain functional differences compared to adult T cells, including the apparent tendency in favor of Th2 responses, which differences appear to reflect in large part the naïve status of the neonatal immune system and lack of exposure to bacteria and other microbes, which encourage Th1 responses, prior to birth (reviewed in English et al., 2001; Lewis and Wilson, 2001; Prescott et al., 1998; Siegrist, 2001). Antigen Exposure Through Vaccines Central to the safety concerns about multiple childhood immunizations is the question of whether the increasingly complex recommended schedule of immunizations overloads a child’s immune system. That is, have there been quantitative or qualitative changes in the antigens to which a child is exposed through
OCR for page 40
Immunization Safety Review: Multiple Immunizations and Immune Dysfunction vaccines that lead to an inability of the immune system to respond appropriately? A related question involves the “hygiene hypothesis.” These two issues are reviewed here, prior to the review of evidence regarding possible adverse health effects—specifically infection, autoimmune disease, or allergy—of multiple immunizations on the developing immune system. A vaccine directed against a single disease can contain one antigen or can contain multiple antigens, each of which can have multiple epitopes. For example, the polio vaccine has always been directed against three strains of poliovirus. The pneumococcal polysaccharide vaccine (recommended for children older than 24 months of age and for adults) contains antigens for 23 distinct strains of pneumococcal bacteria, and the pneumococcal polysaccharide-protein conjugate vaccine (for children between 2 and 23 months of age) contains antigens for seven distinct strains of pneumococcal bacteria. With other vaccine products, such as the DTaP vaccine or the MMR vaccine, a single inoculation (or “shot”) is directed against several different diseases. Some vaccines are much simpler. For example, the vaccine directed against the hepatitis B virus contains only one protein antigen. Quantitative Considerations Calculations presented to the committee (Kollman, 2001; Offit et al., 2002) suggest that the number of antigens contained in the complete set of vaccines that comprise the recommended childhood immunization schedule has decreased over the past 20 to 30 years, despite the increased number of vaccines and vaccine doses administered. The removal of two vaccines from the schedule account for this decrease. First, routine use of the smallpox vaccine was discontinued in the United States in 1971; the World Health Assembly certified the elimination of wild-type smallpox in May 1980 (CDC, 2001c). The smallpox vaccine contained approximately 200 distinct and potentially antigenic elements. Second, the DTwP vaccine—the whole-cell pertussis (wP) vaccine generally given in combination with diphtheria (D) and tetanus (T)—was replaced by an acellular vaccine DTaP, the first of which was approved by the FDA in 1991. The whole-cell vaccine contained approximately 3,000 distinct and potentially antigenic components, whereas the acellular vaccine contains only 2–5 antigens. As of 1997, the acellular pertussis vaccine is the vaccine of choice in the United States, although the whole-cell preparation is still used elsewhere. The vaccines added to the immunization schedule over the past 20 years have relatively few antigens. For example, the hepatitis B vaccine, a genetically engineered product, contains only one antigen, and the Haemophilus influenzae type b (Hib) vaccine contains only two. The varicella vaccine, a live viral vaccine, contains approximately 70 antigens (see Table 1). Thus, with the elimination of smallpox vaccine and the changeover to the acellular pertussis vaccine, the total number of immunogenic proteins or polysaccharides in childhood
OCR for page 41
Immunization Safety Review: Multiple Immunizations and Immune Dysfunction vaccines has decreased to a level well below that of the vaccines given widely even as recently as 1980 (Kollman, 2001; Offit et al., 2002). Certain caveats must be made regarding these calculations. First, they rely on counting numbers of unique molecules (e.g., proteins) in smallpox and whole cell B. pertussis vaccines—some of which may not be antigenic and others of which contain multiple epitopes to which the immune system responds. The calculations also do not address the effects of changes in the presence or absence of contaminating proteins. For example, the use of antibiotics, growth media, animal proteins, or carrier proteins could alter these preliminary calculations. In addition, there is no attempt to consider inter- or intra-manufacturer differences in vaccine preparations. The other side of the quantitative question regarding antigen load is whether infants are capable of responding adequately to the antigens presented by vaccines. Adult humans have a T cell receptor repertoire (the numbers of unique T cell receptors and thus the number of different epitopes to which the T cells of an individual could respond) of ~2.5×107 (Arstila et al., 1999). Although the numbers of different T cell receptors present in human neonates has not been determined directly, their diversity has been shown by several groups to be similar to that of adults. Thus, the range of different epitopes that human neonates can recognize is almost certainly >107. The diversity of antigens to which B cells can make specific antibodies is thought to be even greater, and although there are some qualitative differences from adults, it appears that the overall diversity of antigens to which B cells can respond is similar to adults by 6–8 weeks of age in humans (English et al., 2001; reviewed in Lewis and Wilson, 2000; Marolleau, 1998). This is the basis TABLE 1 Number of Immunogenic Proteins and Polysaccharides Contained in Vaccines Over the Past 40 Years 1960 1980 2000 Vaccine Protein Vaccine Protein Vaccine Protein Smallpox ~200 Diphtheria 1 Diphtheria 1 Diphtheria 1 Tetanus 1 Tetanus 1 Tetanus 1 WC-Pertussis ~3000 AC-Pertussis 2–5 WC-Pertussis ~3000 Polio 15 Polio 15 Measles 10 Measles 10 Polio 15 Mumps 9 Mumps 9 Rubella 5 Rubella 5 Total ~3217 Hib 2 Total ~3041 Varicella 69 Pneumococcus 8 Hepatitis B 1 Total 123–126 SOURCE: Adapted from Offit et al., 2002 NOTE: WC-Pertussis=whole cell pertussis, AC=acellular pertussis
OCR for page 42
Immunization Safety Review: Multiple Immunizations and Immune Dysfunction for the notion that human infants have the capacity to respond to the substantial number of foreign molecules (e.g., bacterial antigens) to which they are exposed shortly after birth. This is consistent with the theoretical estimates presented to the committee, which suggest that the capacity of the infant’s immune system is at least 1000 times greater than that maximally required to respond to vaccines (Kollman, 2001; Offit et al., 2002). It is the judgment of several scientific groups, including the Immunization Safety Review Committee, that the antigen load of the recommended childhood immunization schedule has decreased, not increased, in the last 20 years or so and that the infant immune system has an adequate capacity to respond to that number of antigens. Qualitative Considerations In considering whether the additional exposure to vaccine antigens might “overwhelm” the infant immune system, reference has been made to the fact that the fetus moves from a sterile environment in the womb into the birth canal and outside world that is coinhabited by an almost infinite array of microorganisms (IOM, 1994). Within hours, the newborn’s skin and upper respiratory and intestinal tracts are colonized by a variety of bacteria and fungi, and exposure to viruses begins. Thus, the baseline exposure to microbial antigens by an infant is very large. The normal infant develops a “commensal” relationship with these bacteria and fungi, almost always without preceding expression of overt disease—a sort of truce between the host and microbe that allows the microbe to colonize but not invade. During this process, the immune system is stimulated by these exposures, as illustrated by the presence of detectable antibody and lymphocyte responses to organism-specific antigens. Antibodies against the common pathogens H. influenzae and pneumococci has also been demonstrated in infants not recognized to have had disease caused by these bacteria (Anderson et al., 1972; Gray et al., 1981; Sell et al., 1973). Moreover, genetic immunodeficiency diseases represent “experiments of nature” that show that abnormality of any single component of the host defense system, including antibody and lymphocyte function, can result in serious, often lethal disease caused by pathogens or by one or another of the commensal organisms. Thus a vigorous immune response is required to protect the human infant against infection by a broad variety of organisms that have the potential to cause disease, and the infant must be able to mount this response consistently and repeatedly. Within this context, it seems unlikely that immunizations constitute a significant departure from the magnitude of the antigenic challenges endured under natural circumstances by any normal infant. Over the course of several decades, the antigen load presented to the developing immune system has undergone significant qualitative changes,
OCR for page 43
Immunization Safety Review: Multiple Immunizations and Immune Dysfunction particularly in the context of the total antigen exposures during infancy and childhood. Approximately a decade ago, researchers interested in the changing epidemiology of several diseases began formulating the “hygiene hypothesis,” which has generated an extensive descriptive and research literature (e.g., Rook, 2000; Wills-Karp et al., 2001). Fundamentally, the hygiene hypothesis suggests that the increasingly aseptic environment in which children in developed countries are reared has led to changes in the development, or maturation, of the infant immune system. Many ecological analyses correlate the rise of allergic and autoimmune diseases in many parts of the world with increased economic development (Rook and Stanford, 1998). Epidemiological literature in support of the hygiene hypothesis includes findings of a negative correlation between risk for allergic diseases and a host of factors that would increase a child’s exposure to bacteria and other infectious agents. These risk factors include, for example, the number of older siblings, the presence of pets, infections through the fecaloral route, and rural living. Changes other than in hygienic behavior, such as increases in environmental pollutants, have also occurred in the developed world and may contribute to the changing epidemiology of some diseases. The proposed explanation for an immune system role in these epidemiological observations is that early exposure to infectious diseases and environmental microbes “shapes” the developing immune system toward a Th1-cell responsiveness, which is generally considered a protective immune response (i.e., to host defense against intracellular pathogens and allergy). Eliminating these early exposures through hygienic practices and altered behaviors is thought to predispose the immune system toward a Th2 cell responsiveness, which is associated with allergy. This “skewing” or “biasing” of the immune system as a result of the elimination of many kinds of antigen exposures, some theorize (Rook, 2001), is exacerbated by exposure to vaccines, many of which evoke a Th2 response instead of the Th1 response that would be generated by wild-type infections with the diseases that the vaccines prevent. The most recent refinement of the biological mechanisms proposed to explain the hygiene hypothesis looks beyond the idea of a simple Th1-Th2 imbalance into the realm of regulatory cell imbalance (Rook, 2001; Wills-Karp et al., 2001). Under this scenario, a major contributor to altered immune responses is a decrease in T regulator cells, along with the alteration in T effector cells (Th1 or Th2). Not yet clear is the role immunizations may have in directly altering development of the immune system, or the relative contribution of vaccine-related changes in the context of the hygiene hypothesis. Vaccine-induced immune responses may, however, differ from those resulting from wild-type infection because of differences in context, including differences in their timing, either in terms of age at exposure or of the sequence of antigen exposure. For example, through immunization, many American children currently mount a simultaneous immune response to diphtheria, tetanus, pertussis, hepatitis B, Haemophilus influenzae type b, and three strains of poliovirus three times during the first 6
OCR for page 44
Immunization Safety Review: Multiple Immunizations and Immune Dysfunction months of life. It is unlikely that the timing of this vaccine-induced immune response would have been mimicked under conditions of nonvaccination. In any case, the number of infections prevented by vaccines is actually quite small compared with the total number of infections prevented by other hygienic interventions, such as clean water, food, and living conditions. And, although it is true that the developing immune system is frequently bombarded with many antigens at one time, most of the antigens do not pose a threat inherent to the infant. Most certainly, the route of antigen exposures through vaccines—that is, an injection rather than a respiratory or gastrointestinal exposure—is different than what occurs in wild-type infection. Actually, the history of vaccine development shows that the immune response to a vaccine is sometimes devastatingly different from the response to wild-type infection. Early experience with killed-virus vaccines directed against measles and respiratory syncytial virus (RSV) saw the appearance of atypical and virulent disease in vaccinated individuals that were subsequently infected with wild-type virus. In the 1960s, some children developed an atypical form of measles after receiving the killed measles virus vaccine (Fulginiti and Helfer, 1980). Atypical measles is described as a delayed, severe hypersensitivity reaction (Krause et al., 1978; Redd et al., 1999). Symptoms of high fever, headache, abdominal pain, myalgia, and cough (Redd et al., 1999) are followed in 48–72 hours by the appearance of a maculopapular, pruritic rash on the extremities that spreads inward toward the trunk and may become vesicular, purpuric, or petechial (Brodsky, 1972). Patients become severely ill during the first few days of illness, but atypical measles is self-limited and resolves in 7–14 days (Brodsky, 1972).There was only one report of a possible fatality among cases seen in the 1960s following use of the inactivated measles vaccine (Redd et al., 1999). Children with atypical measles were found to lack the antibody to the measles virus F protein, which is responsible for the virus’s hemolytic and cell fusion properties (Annunziato et al., 1982; Redd et al., 1999). In contrast, the H, or hemagglutin protein, the other measles virus surface protein, was found in the sera of the ill patients. This indicated that children given the killed-virus vaccine formed antibody against the H protein even though they did not do so against the F protein (Annunziato et al., 1982). There was also a suggestion of an exaggerated cellular immune response to measles antigens in patients (Redd et al., 1999), although more recent studies in rhesus monkeys suggest that the induction of humoral and CD4 T cell-mediated immunity but not cytotoxic T cells directed against viral antigens may be an important factor in the adverse response to subsequent infection with wild-type virus which resulted in the production of extremely high levels of circulating antibody (Polack et al., 1999; Redd et al., 1999). Years after being vaccinated with killed virus, patients who contracted measles were still developing a clinical illness that, aside from the
OCR for page 45
Immunization Safety Review: Multiple Immunizations and Immune Dysfunction initial symptoms, was quite different from regular measles (Annunziato et al., 1982). A similar effect was seen when a vaccine against RSV infection—the leading cause of lower respiratory tract illness among infants and children—was tested. Susceptible children were vaccinated with a concentrated, formalininactivated, adjuvant-enhanced vaccine. Children who received this vaccine initially developed high levels of neutralizing and complement-fixing antibody. However, upon exposure to wild-type RSV, they developed more severe infections than unvaccinated children did (CDC, 2000a; IOM, 1985). Thus, the experimental RSV vaccine was never used clinically. Atypical measles syndrome and the severe RSV infections are two disturbing consequences of immunization. These two examples resulted from an incomplete understanding of the immune system’s response to the live and killed form of these viruses. Thus, it is clear that the response to vaccines cannot be predicted, even with caution and reserve and therefore should be assessed carefully. Autoimmune and Allergic Diseases Autoimmune Disease Collectively, diseases of autoimmunity affect 3 to 5 percent of the population in the United States. (Jacobson et al., 1997).Autoimmune diseases are mediated by T cell and/or T cell-dependent B cell responses directed against self-antigens, and the T cell responses in most autoimmune diseases are dominated by interferon-γ producing CD4 T cells, commonly referred to as Th1 T cells (Marrack et al., 2001). An autoimmune process can target individual organs, such as the brain and spinal cord in multiple sclerosis, or can operate throughout the body, as in systemic lupus erythematosus. For this report, the committee focused on the autoimmune form of diabetes, referred to as type 1a diabetes. Type 1b refers to diabetes associated with an idiopathic loss of insulin secretion. Type 2 diabetes is not associated with destruction of insulin-secreting pancreatic islet cells. Type 1 diabetes has frequently been referred to as “childhood” or insulin-dependent diabetes, while type 2 diabetes has been referred to as “adult-onset” diabetes. It is now recognized that onset of either form of diabetes can occur at any age. Type 1a diabetes results from the immunological destruction of pancreatic islet ß cells (Atkinson and Eisenbarth, 2001). (The destruction of the islet cells in type 1b diabetes is idiopathic). The beta cells produce insulin, which the body requires to process glucose. Symptoms of type 1 (a and b) diabetes include increased thirst and urination, constant hunger, weight loss, blurred vision, and extreme fatigue. Once diagnosed, the condition can be treated with regular injection of supplemental insulin, but various long-term complications (e.g., diabetic retinopathy, kidney failure, vascular disease) are common. The
OCR for page 46
Immunization Safety Review: Multiple Immunizations and Immune Dysfunction development of clinical disease is preceded by an asymptomatic period of progressive islet destruction that may last for many years. As understanding of type 1a diabetes increases, the preclinical period may provide a window for interventions that can delay or prevent clinical onset (Knip, 1997). The etiology of type 1a diabetes and other autoimmune diseases is multifactorial, involving genetics and environmental exposures. Genetic susceptibility, arising from combinations of multiple genetic factors, appears to be a necessary but not sufficient risk factor for the disease. Monozygotic twins of persons with type 1a diabetes are at increased risk compared with other family members, but as few as 23 percent of these twins developed the disease in one report (Abiru and Eisenbarth, 2000). Although the disease can cluster in families, more than 80 percent of type 1 diabetes cases are reported to occur in persons with no family history of the disease (Dorman et al., 1995). Certain genetic factors may also provide protection from type 1 diabetes. Environmental factors are thought to serve as triggers or promoters of the autoimmune process in genetically susceptible individuals. In particular, dietary and viral exposures have been suspected in type 1 diabetes. Some studies found that early exposure to cow’s milk was associated with increased risk (Gerstein, 1994), as was breastfeeding for less than 3 months (Gerstein, 1994). Newer prospective studies, however, have found no association with these factors (Graves et al., 1999; Hummel et al., 2000; Norris et al., 1996). Congenital rubella syndrome (CRS) shows a clear association with type 1 diabetes, with about 20 percent of CRS patients in the United States also having diabetes (Menser et al., 1978; Rubinstein et al., 1982). Studies of Coxsackie B virus infections have produced conflicting evidence regarding their possible contribution to type 1 diabetes (Robles and Eisenbarth, 2001). Atkinson and Eisenbarth (2001) described a model of progression to clinical disease that depends not on exposure to a single triggering environmental agent but rather on the cumulative effect of various exposures over time. Worldwide, estimates of the incidence of type 1 diabetes4 in children under 14 years of age range from 0.1 per 100,000 in parts of China and Venezuela to 36.8 per 100,000 in Sardinia and 36.5 per 100,000 in Finland (Karvonen et al., 2000). As reported by Karvonen and colleagues (2000), estimated incidence for the early 1990s for United States locations range from 11.7 per 100,000 in Chicago to 17.8 per 100,000 in Allegheny County, Pennsylvania. In most populations, incidence is highest in the oldest age group (10–14 years). The disease is also diagnosed in adults, but incidence data are limited. Data from Rochester, Minnesota, for 1945–1969 suggest an incidence rate of 9.2 per 100,000 among persons age 20 and older (Melton et al., 1983). Rates have generally been lower in more tropical countries and higher in populations of European origin. These patterns may be related to differences 4 Most population-based studies do not distinguish between types 1a and 1b diabetes.
OCR for page 47
Immunization Safety Review: Multiple Immunizations and Immune Dysfunction among racial and ethnic groups in the distribution of genetic risk factors or to the differences in exposure to environmental factors (Atkinson and Eisenbarth, 2001). In any case, the worldwide incidence of type 1 diabetes appears to be increasing 3% a year, and the rate of increase is greatest where incidence has been low (Onkamo et al., 1999). Data from Europe also indicate that incidence is increasing more rapidly among the youngest children (ages 0–4 years) (EURODIAB, 2000). Allergy Allergy is responsible for a variety of acute and chronic health problems, including anaphylaxis and allergic rhinitis, asthma, and eczema. These conditions reflect an overreaction of the immune system to allergens—normally harmless environmental agents such as pollens, dust mites, insect venom, and certain foods—that can be encountered through inhalation, ingestion, injection, or skin contact. Under certain circumstances, exposure to an allergen primes the immune system for hypersensitivity reactions involving allergen-specific IgE antibodies and Th2 cells. The committee focused its attention on allergic asthma for this report. Characteristic symptoms of asthma are episodes of shortness of breath, coughing, wheezing, and chest tightness. These symptoms reflect an acute bronchial hyperresponsiveness to specific allergens and other environmental factors, and a chronic inflammation of the airways (Busse and Lemanske, 2001; IOM, 2000; Kay, 2001; Parham, 2000). The acute response involves activation of mast cells in the lower airways and their release of histamine, cytokines, and other molecules. These mast-cell mediators induce accumulation of fluid, secretion of mucus, and contraction of the smooth muscle around the airways. A “late phase” response includes persistent or recurrent bronchial constriction and infiltration of airway tissue by inflammatory cells. The inflammation can produce temporary or permanent tissue damage. Exposures to allergens and other environmental factors are known to induce new episodes of asthma when the disease is established, but the underlying factors that account for the development of this type of hyperresponsiveness are not fully understood. Several genetic factors may combine in various ways to establish susceptibility, but they remain poorly defined (Barnes, 2000; Kay, 2001). Environmental exposures may also influence the development of asthma, but current evidence is mixed for exposure such as cockroach allergens or cat dander (e.g., IOM, 2000; Lau et al., 2000; Litonjua et al., 2001; Peat et al., 1993). The presence of older siblings or attendance at day-care (factors that may be markers for the nature or timing of certain exposures) has shown protective effects (Ball et al., 2000). Early exposure to certain viral infections also has shown protective effects (Illi et al., 2001; Openshaw and Hewitt, 2000), but some
Representative terms from entire chapter: