Cover Image

PAPERBACK
$43.50



View/Hide Left Panel

Appendix D-16
The Prospects for Immunizing Against Streptococcus Group A

DISEASE DESCRIPTION

The group A streptococcus (GrAS) is a ubiquitous microorganism that causes a wide range of human infections, including acute tonsillopharyngitis, sinusitis, otitis, scarlet fever, erysipelas, impetigo, cellulitis, lymphangitis, pneumonia, and endometritis. The impetus for vaccine development, however, stems from the propensity of the organism to elicit the delayed nonsuppurative sequelae of acute rheumatic fever (ARF) and acute post-streptococcal glomerulonephritis (AGN). Because the two sequelae appear to be caused by different GrAS serotypes, and because ARF appears to present a much greater worldwide health threat than does AGN, this brief review will focus exclusively on the prospects for prevention of ARF. However, the technology of vaccine production, once perfected, would be directly applicable to prevention of AGN.

Essential for the development of ARF is an antecedent GrAS infection of the upper respiratory tract. This scenario differs from AGN, which may follow either pharyngeal or cutaneous streptococcal infection. ARF usually manifests itself 1 to 4 weeks (median, 19 days) after the GrAS infection, at a time when all symptoms of the acute bacterial process have abated. Only a small proportion of individuals experiencing such an infection will, however, develop ARF. Depending on the epidemiologic setting, this proportion varies from about 3 percent to less than 1 percent of patients experiencing an untreated GrAS infection.

The five major clinical manifestations of ARF are well known because of their inclusion in the modified Jones criteria for diagnosis of this disease. These manifestations are: carditis, polyarthritis, Sydenham’s chorea, subcutaneous nodules, and erythema marginatum. The clinical presentation of the disease is quite variable, because the individual manifestations may occur singly or in combination. The degree of attendant systemic toxicity may likewise vary greatly. An acute attack

The committee gratefully acknowledges the efforts of A.L.Bisno and E.H.Beachey, who prepared major portions of this appendix, and the advice and assistance of S.Dodu and H.B.Houser. The committee assumes full responsibility for all judgments and assumptions.



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 338
New Vaccine Development: Establishing Priorities, Volume II, Diseases of Importance in Developing Countries Appendix D-16 The Prospects for Immunizing Against Streptococcus Group A DISEASE DESCRIPTION The group A streptococcus (GrAS) is a ubiquitous microorganism that causes a wide range of human infections, including acute tonsillopharyngitis, sinusitis, otitis, scarlet fever, erysipelas, impetigo, cellulitis, lymphangitis, pneumonia, and endometritis. The impetus for vaccine development, however, stems from the propensity of the organism to elicit the delayed nonsuppurative sequelae of acute rheumatic fever (ARF) and acute post-streptococcal glomerulonephritis (AGN). Because the two sequelae appear to be caused by different GrAS serotypes, and because ARF appears to present a much greater worldwide health threat than does AGN, this brief review will focus exclusively on the prospects for prevention of ARF. However, the technology of vaccine production, once perfected, would be directly applicable to prevention of AGN. Essential for the development of ARF is an antecedent GrAS infection of the upper respiratory tract. This scenario differs from AGN, which may follow either pharyngeal or cutaneous streptococcal infection. ARF usually manifests itself 1 to 4 weeks (median, 19 days) after the GrAS infection, at a time when all symptoms of the acute bacterial process have abated. Only a small proportion of individuals experiencing such an infection will, however, develop ARF. Depending on the epidemiologic setting, this proportion varies from about 3 percent to less than 1 percent of patients experiencing an untreated GrAS infection. The five major clinical manifestations of ARF are well known because of their inclusion in the modified Jones criteria for diagnosis of this disease. These manifestations are: carditis, polyarthritis, Sydenham’s chorea, subcutaneous nodules, and erythema marginatum. The clinical presentation of the disease is quite variable, because the individual manifestations may occur singly or in combination. The degree of attendant systemic toxicity may likewise vary greatly. An acute attack The committee gratefully acknowledges the efforts of A.L.Bisno and E.H.Beachey, who prepared major portions of this appendix, and the advice and assistance of S.Dodu and H.B.Houser. The committee assumes full responsibility for all judgments and assumptions.

OCR for page 338
New Vaccine Development: Establishing Priorities, Volume II, Diseases of Importance in Developing Countries of rheumatic fever, although often disabling, runs its course in a matter of weeks, and death during such an attack is rare nowadays in the developed world. When fatalities do occur, they are due to intractable rheumatic carditis. Detailed clinical descriptions of ARF are available in numerous reviews (Bisno, 1982, 1985; Stollerman, 1975; Whitnack and Bisno, 1980). From a public health standpoint, the intense concern with ARF stems from its propensity to inflict permanent heart damage. Such damage usually takes the form of chronic, deforming rheumatic valvulitis leading to insufficiency and stenosis of the mitral or aortic valves or both and occasionally of the valves of the right heart as well. Moreover, patients who have suffered a single attack of ARF are highly susceptible to recurrent attacks following immunologically significant GrAS upper respiratory infections. As the rheumatic attacks tend to be mimetic, individuals who experienced carditis with the first attack are likely to experience progressive heart damage with succeeding episodes. The long-term prognosis of ARF is closely correlated with the cardiac status during the acute attack. This was shown most conclusively in the joint United Kingdom/United States collaborative study (United Kingdom and United States Joint Report, 1965), wherein some 494 children under the age of 16 were studied; about 70 percent of these were followed up over many years. Among patients free of carditis during their acute attack, only 6 percent had residual heart disease. Patients with no preexisting heart disease who experienced mild carditis during their acute attack (i.e., apical systolic murmur without pericarditis or heart failure) had a relatively good prognosis in that only about 30 percent had heart murmurs 19 years later. About 40 percent of subjects with apical or basal diastolic murmurs and 70 percent of subjects with congestive heart failure or pericarditis or both during their acute attacks had residual rheumatic heart disease. The prognosis was worse in patients with preexisting heart disease and in those who experienced recurrent attacks of ARF. Thus, as a general rule, patients who do not experience carditis during an initial attack of ARF and who are protected from recurrent rheumatic attacks will not go on to develop rheumatic heart disease. Individuals with “pure” chorea represent an exception to this rule, because 25 percent of them may go on to develop rheumatic heart disease. Although the figures vary widely, in most modern studies about 50 percent of patients diagnosed as having ARF experience some carditis during the acute attack. One might estimate that some 33 to 50 percent of such patients will be left with residual rheumatic heart disease of varying severity. Thus, perhaps one-quarter of all ARF patients develop chronic cardiac involvement. In many of these patients, the involvement is characterized by severe valvulitis with chronic congestive heart failure. A sizeable number of patients eventually require valve replacement or die from the effects of rheumatic heart disease.

OCR for page 338
New Vaccine Development: Establishing Priorities, Volume II, Diseases of Importance in Developing Countries PATHOGEN DESCRIPTION Streptococci are spherical or ovoid bacteria that grow in pairs or chains of varying lengths. They are gram-positive, non-spore-forming, catalase-negative, and ordinarily nonmotile. They have complex and variable nutritional requirements. Taxonomically, these organisms belong to the family Sreptococcaceae, genus Streptococcus, of which there are 21 identified species. When streptococci are cultivated on blood agar plates, they may produce complete (beta) hemolysis, partial (alpha) hemolysis, or no (gamma) hemolysis. More precise differentiation of streptococci can be made by immunologic means. The organisms may be separated into serogroups by means of antigenic differences in the cell wall carbohydrates or teichoic acids. The group A streptococcus, also known as Streptococcus pyogenes, is facultatively anaerobic, beta-hemolytic (with very rare exceptions), and contains in its cell wall a group-specific carbohydrate composed of rhamnose and N-acetyl-glucosamine. There are a number of other important substances expressed on the streptococcal surface. These include lipoteichoic acid, the principal ligand responsible for adherence of GrAS to epithelial surfaces, and a variety of proteins designated as M, T, and R. Of these, by far the most important is M protein. M protein is the major virulence antigen of GrAS. Strains rich in this protein are resistant to phagocytosis by polymorphonuclear leukocytes, multiply rapidly in fresh human blood, and are capable of initiating disease. Strains lacking M protein are avirulent. Group A streptococci may be divided into serotypes on the basis of antigenic differences in M protein molecules. Over 70 such serotypes are currently recognized. Human immunity to streptococcal infection is based on the development of type-specific opsonic antibodies directed against the anti-phagocytic moiety of M protein. In rare instances, cross-protection by antibody to one type against organisms of a heterologous type has been demonstrated. The M protein molecule penetrates the cell wall; this configuration localizes the type-specific antigen on the tips of hair-like fimbrillae protruding from the cell surface. The manner in which M protein exerts its anti-phagocytic effect is under investigation. The protein prevents interaction of the streptococcal cell wall with complement components, an effect that is enhanced by the ability of M protein to precipitate fibrinogen directly onto the bacterial surface. This protective effect is nullified by the presence of adequate concentrations of type-specific antibody. T protein serves as the basis for a subsidiary typing system for GrAS. Another important somatic component of the streptococcus is the hyaluronic acid capsule. This capsule is also anti-phagocytic and serves as an auxiliary virulence factor. For many years, it has been recognized that the ability to initiate AGN is limited to strains of certain GrAS serotypes (e.g., types 1, 4, 12, 49, 55, and a few others). Although this distinction between nephritogenic and nonnephritogenic GrAS is well established, there has been considerably more controversy over the issue of whether GrAS vary

OCR for page 338
New Vaccine Development: Establishing Priorities, Volume II, Diseases of Importance in Developing Countries in their rheumatogenic potential as well. A growing body of evidence now strongly supports the concept that a limited number of M protein serotypes account for the great majority of ARF cases in any particular epidemiologic setting (Bisno, 1980; Bisno et al., 1970). M type 5 appears to be the most highly rheumatogenic type, while others of the classic pharyngitis serotypes, such as 3, 6, 14, 18, 19, and 24, are also strongly implicated. This recognition of the variable rheumatogenicity of GrAS strains has given new impetus to laboratory analysis of the antigenic composition of GrAS and in addition provides important new insights into the most effective strategies for vaccine development and implementation. HOST IMMUNE RESPONSE During the course of human infection, the host mounts both humoral and delayed immune responses to a wide variety of streptococcal somatic and extracellular products. Patients with ARF are apparently hypersensitive to streptococcal antigens, because their mean antibody titers are significantly higher than similar titers among individuals with uncomplicated GrAS infection. For the purposes of this review, only humoral immune responses to GrAS extracellular products and M protein will be considered. During the course of growth in vivo or in vitro, GrAS elaborates a multitude of extracellular products, many of which are antigenic. Measurement of antibodies to certain of these extracellular products has proved of great value, both to the clinician and the epidemiologist. The two most commonly used antibody tests clinically are the anti-streptolysin O (ASO) and anti-deoxyribonuclease B (anti-DNAse B) tests. Other tests used from time to time are the anti-hyaluronidase (AH), anti-NADase, and anti-streptokinase assays. One or more of these antibodies are elevated above the usual upper limits of normal in virtually all patients with ARF. The modified Jones criteria for the diagnosis of ARF require, in addition to a compatible clinical picture, evidence of recent streptococcal infection. Such evidence may be provided by a positive throat culture for GrAS, a classic scarlatinal rash, or, most frequently, an elevated or rising titer of antibodies to streptococcal extracellular products. Indeed, ARF rarely if ever occurs without evidence of an immunologically significant GrAS infection, as judged by elevation of one or more of these antibody titers. The appearance of these antibody titers may be delayed or their magnitude suppressed by prior antibiotic therapy. Type-specific anti-M antibody responses are more difficult to measure than ASO or anti-DNAse B titers. While the latter are assayed by simple neutralization tests of the action of streptolysin O or DNAse B, the former require precise and cumbersome assays, such as mouse protection or opsonophagocytic tests. Type-specific anti-M responses occur following untreated GrAS pharyngeal infections but are markedly suppressed by effective antibiotic therapy. The immune response is type-specific, protective, and quite durable. Anti-M antibodies protect experimental animals from lethal streptococcal challenge and,

OCR for page 338
New Vaccine Development: Establishing Priorities, Volume II, Diseases of Importance in Developing Countries although the data are limited, appear to confer protection against reinfection in humans as well. Because M protein is the sole streptococcal antigen known to elicit protective antibodies, modern studies of streptococcal immunization have focused on development of a safe and immunogenic M protein preparation. Details of such studies are outlined below in the section on prospects for vaccine development. Although the pathogenesis of ARF remains unknown, most authorities believe that the disease results from host immune responses to certain streptococcal antigens that share antigenic determinants with human host tissues. For this reason, any putative M protein vaccine should consist of that portion of the M protein macromolecule that is responsible for eliciting type-specific immunity, while at the same time being free from antigens associated with M protein (so-called M associated proteins or non-type-specific M antigens) that might cross-react with host tissues. DISTRIBUTION OF DISEASE Geographic Distribution Exact data on the geographic distribution of ARF and rheumatic heart disease are not available. However, the worldwide patterns of occurrence are clear. During this century, there has been a sustained and profound decline in rheumatic fever incidence in the developed countries of North America and Western Europe. Other highly developed countries, such as Japan, are also experiencing a marked drop in disease incidence. In contrast, ARF incidence is not declining, and may even be increasing, in many of the developing areas of the world. Such areas include the Indian subcontinent, the Arab countries of the Middle East, selected areas in sub-Saharan Africa and South America, and certain highly susceptible populations, such as the Maoris of New Zealand. The incidence of ARF tends to be highest in the thickly congested, low-income areas of the world’s major cities. Disease Burden Estimates Although figures on the incidence of rheumatic fever are difficult to obtain and often unreliable, there is no doubt that rheumatic fever remains one of the major causes of cardiovascular morbidity and mortality in the developing nations of the world. As noted above, the disease is rampant in the Indian subcontinent, the Middle East, and many countries of Africa and South America. In Sri Lanka in 1978, for example, the morbidity rate of ARF was 47 per 100,000 population and over 140* for the 5 to 19 years age group (World Health Organization, *   All rates in this section are expressed as cases per 100,000 population at risk, unless otherwise indicated.

OCR for page 338
New Vaccine Development: Establishing Priorities, Volume II, Diseases of Importance in Developing Countries 1980). At Rangpur Medical College Hospital in Bangladesh and Rangoon General Hospital in Burma, acute and chronic rheumatic heart disease account for about one-third of cardiac admissions and 2 percent of all hospital admissions. In India, the prevalence of rheumatic heart disease among children 5 to 15 years of age has been estimated at 600 per 100,000. A relatively recent aspect of the epidemiology of rheumatic fever is the recognition of its frequent occurrence in subtropical and tropical regions of the world. Prior to 1950, there was little or no awareness of the disease as a major public health problem in such areas (Markowitz, 1981). Although adequate longitudinal data are unavailable, ARF incidence actually may have increased substantially in the tropics during the middle of this century, a trend perhaps due in part to major population growth and increasing urbanization. In contrast, morbidity and mortality of ARF and rheumatic heart disease have declined markedly in western Europe and North America, a decline that appears to have begun prior to the antibiotic era and is best documented in the Scandinavian data (Stollerman, 1975). While similar sequential data are lacking for most cities in the United States, one has only to query older physicians or to observe the closing of the famous rheumatic fever sanatoriums of the past to perceive how much the rate has fallen in certain areas. Detailed epidemiological surveys of ARF incidence were conducted during the 1960s in several U.S. cities. In the borough of Manhattan (Brownell and Bailen-Rose, 1973), the overall annual rate among school-aged children was 61 per 100,000; the highest rates, 78 to 79, occurred in the most congested Puerto Rican neighborhoods, and the lowest rate, 23, in the district with the largest proportion of white middle-class families. Annual ARF incidence rates among school-aged children in Baltimore (Gordis et al., 1969) and Nashville (Quinn and Federspiel, 1974) were in the range of 24 to 34. In all these studies, rates for nonwhites were considerably higher than those for whites. Studies conducted in the late 1970s have shown remarkably low rates of ARF incidence in the United States. For example, a careful survey of disease incidence in Memphis, Tennessee (Land and Bisno, 1983), for the years 1977 to 1981 indicated an incidence among 5 to 17 year olds of only 1.88 per 100,000 population per annum. Moreover, the ARF incidence among white school children dwelling in suburban and rural portions of Memphis was only 0.5. Since publication of the Memphis data, similar rates have been reported from Fairfax County, Virginia; Baltimore, Maryland; Los Angeles, California; and the state of Rhode Island. A recent publication suggests that the developing world may also experience a decline in the ARF and RHD as the standard of living and medical care improve (Lancet, 1985). Derivation of Estimates The estimate of the disease burden from GrAS developed below focuses on acute rheumatic fever and its consequences; acute glomerulonephritis

OCR for page 338
New Vaccine Development: Establishing Priorities, Volume II, Diseases of Importance in Developing Countries is not included since it is not the target of the present vaccine development efforts (the strains that cause it differ from those that cause ARF, and it appears to be a much smaller worldwide health threat). The contribution of minor GrAS infections to the total disease burden has not been considered in this analysis. Markowitz (1981), in reviewing the epidemiology of rheumatic fever, suggested that the incidence of ARF and RHD in developing countries approximated that found in western Europe and the United States 70 to 80 years ago, that is, 150 cases per 100,000 population. However, on the basis of recent studies in developing countries, the annual incidence of ARF is thought to range from 20 to 100 cases per 100,000 population, with 60 as an estimated average for the developing world (Dodu, personal communication, 1985). For any country or region, the rate would probably be higher in urban areas and lower in rural. This rate would yield a total of 2.157 million cases of ARF among the population of the developing world (about 3.595 billion). Markowitz (1981) estimated that 15 percent of all cases of rheumatic fever in developing countries are in the under 5 years age group. On the basis of the age distribution for ARF in developed countries (Quinn, 1982) and information supplied by Dodu (personal communication, 1985) that in developing countries the peak of illness is at a lower age (about 7 to 9 years), it is estimated that 80 percent of total ARF occurs in the 5 to 14 years age group and that only 5 percent occurs in the 15 to 59 years age group. Markowitz (1981) reported that for a number of developing countries, carditis occurs in 60 to 84 percent of ARF patients. Recent estimates suggest that the figure is in the range of 40 to 50 percent (Dodu, personal communication, 1985). For the distribution of ARF cases between the acute morbidity severity categories used in this analysis, it is assumed that 60 percent fall into category C, 30 percent into category B, and 10 percent in category A. In developing countries, annual mortality from ARF currently ranges from about 0.1 death per 100,000 population to about 1.2 per 100,000 with an average of 0.5 deaths per 100,000 population (Dodu, personal communication, 1985). Thus, the total number of ARF-associated deaths in the developing world would be about 17,975. These are assumed to be distributed among age groups in a similar fashion to ARF cases (see above). RHD represents a common sequela of ARF. For the purposes of these calculations, it is assumed that in developing countries an average of 50 percent of ARF cases develop RHD. Usually, RHD develops within about 5 years of ARF. Hence, most cases of RHD would become manifest in early adolescence (if the peak of disease in developing countries is assumed to be prior to 10 years of age). Cases of RHD are therefore assumed to be distributed between the 5 to 14 and 15 to 59 years age groups in the proportions 0.90:0.10. RHD cases are assumed to be distributed among the chronic morbidity severity categories D, E, and F in the proportions 0.2:0.6:0.2, respectively. Cases progressing to severity category F probably die. Current mortality from RHD in developing countries is probably on the order of 1 death per 100,000 population (Dodu, personal communication, 1985). Hence, total annual mortality from RHD in the developing world is estimated to be about

OCR for page 338
New Vaccine Development: Establishing Priorities, Volume II, Diseases of Importance in Developing Countries 35,950. Deaths from RHD are assumed to be distributed equally between the 5 to 14 and 15 to 59 years age groups. Table D-16.1 shows the distribution of ARF and RHD attributable to GrAS derived from adopting these assumptions. VACCINE TARGET POPULATION The peak incidence of ARF is in the 5 to 15 years age group in the developed world (Quinn, 1982), but it is lower in developing countries (Dodu, personal communication, 1985). However, in the developing countries, where a GrAS vaccine would be applied, immunization should be achieved as early as feasible. This is because of the intensity of the streptococcal exposure and because in certain developing countries, severe mitral valvular disease, presumably rheumatic in origin, develops much more precociously than it does among western populations. The exact timing, number of doses, and number of required boosters for a streptococcus-ARF vaccine have yet to be ascertained. Nevertheless, it seems quite likely that such a vaccine, when developed, could be readily incorporated into the WHO Expanded Program on Immunization (WHO-EPI). Already, the WHO has major programs in most developing countries of the world related to prevention of rheumatic heart disease by secondary prophylaxis. Therefore, it would be natural for the organization to shift its emphasis in part or in whole to vaccine implementation, once this mode of prophylaxis becomes available. The vaccine probably should be administered to all children in countries where the ARF risk is high, although it initially could be selectively targeted to pockets of particularly intense ARF occurrence. A second and extraordinarily important target population would be the individual who has had a single attack of ARF. Such individuals are exquisitely susceptible to recurrent attacks, and, as indicated above, such recurrent attacks can lead to severe and life-threatening forms of chronic rheumatic valvulitis. Thus, in the group of patients who have already experienced a single rheumatic attack, we have at once both the highest risk population for ARF and the population in greatest danger of serious sequelae, should the disease recur. Vaccine Preventable Illness* Given a perfect streptococcal-ARF vaccine, a very substantial proportion of the burden of rheumatic fever in the world could be prevented by an effective universal immunization program in early childhood. A large majority of initial attacks that occur in children *   Vaccine preventable illness is defined as that portion of the disease burden that could be prevented by immunization of the entire target population (at the anticipated age of administration) with a hypothetical vaccine that is 100 percent effective (see Chapter 7).

OCR for page 338
New Vaccine Development: Establishing Priorities, Volume II, Diseases of Importance in Developing Countries TABLE D-16.1 Disease Burden: Streptococcus A     Under 5 Years 5–14 Years 15–59 Years 60 Years and Over Morbidity Category Description Number of Cases Duration Number of Cases Duration Number of Cases Duration Number of Cases Duration A Moderate localized pain and/or mild systemic reaction, or impairment requiring minor change in normal activities, and associated with some restriction of work activity 32,355 7 172,560 7 10,785 7     B Moderate pain and/or moderate impairment requiring moderate change in normal activities, e.g., housebound or in bed, and associated with temporary loss of ability to work 97,065 14 517,680 14 32,355 14     C Severe pain, severe short-term impairment, or hospitalization 194,130 21 1,035,360 21 64,710 21     D Mild chronic disability (not requiring hospitalization, institutionalization, or other major limitation of normal activity, and resulting in minor limitation of ability to work)   n.a. 194,130 n.a. 21,570 n.a.   n.a. E Moderate to severe chronic disability (requiring hospitalization, special care, or other major limitation of normal activity, and seriously restricting ability to work)   n.a. 582,390 n.a. 64,710 n.a.   n.a. F Total impairment   n.a. 194,130 n.a. 21,570 n.a.   n.a. G Reproductive impairment resulting in infertility   n.a.   n.a.   n.a.   n.a. H Death, total 2,696 n.a. 31,355 n.a. 17,874 n.a.   n.a.   Acute Rheumatic Fever 2,696 n.a. 14,380 n.a. 899 n.a.   n.a.   Rheumatic Heart Disease   n.a. 16,975 n.a. 16,975 n.a.   n.a.

OCR for page 338
New Vaccine Development: Establishing Priorities, Volume II, Diseases of Importance in Developing Countries and early adolescents would be prevented, and recurrent attacks, which now account for 20 percent or more of all instances of ARF, also would be effectively prevented. There is considerable knowledge of the rheumatogenic potential of various serotypes (see above). However, no vaccine is likely to embrace all rheumatogenic types (there may well be types not yet identified), and because the vaccine could possibly induce some local shifts in the distribution of prevalent streptococcal serotypes, one might anticipate a small number of breakthroughs of rheumatic fever despite adequate immunization. Current vaccine research is focused, however, on the development of pools of small peptides that would protect broadly against a wide variety of streptococcal serotypes. Based on these considerations, it is estimated that 90 percent of the ARF and RHD burden is potentially vaccine preventable. Even a more limited vaccine application, that is, to patients who have already had a single episode of ARF, would decrease rheumatic fever incidence by 20 percent and the incidence of severe carditis by a much greater amount. SUITABILITY FOR VACCINE CONTROL Because of the limitations inherent in currently available preventive measures and because of the devastating consequences of rheumatic fever throughout the developing world, new and innovative approaches to control are required. Clearly, the most appealing of such approaches, both theoretically and practically, would be the development of a safe and effective GrAS vaccine. Such a vaccine would be highly cost-effective because it would obviate the current expenditures in primary and secondary antimicrobial prevention, hospitalization of ARF patients, and other direct and indirect costs associated with the care of large numbers of relatively young individuals with crippling rheumatic cardiac disease. Alternative Control Measures and Treatments There are two existing strategies for the prevention of ARF. The first of these, labeled “primary prevention,” consists of identification of patients with streptococcal pharyngitis followed by effective antimicrobial therapy. Such therapy should be with penicillin, if the patient is not allergic. In high-risk areas of the world, the therapy of choice is a single intramuscular injection of benzathine penicillin G. This serves as treatment for the streptococcal infection and as prophylaxis for ARF. Unfortunately, primary prevention programs often have been difficult to implement in developing countries. in indigent families, little attention is paid to self-limited acute upper respiratory illnesses, and access to medical care is limited. Throat cultures for specific diagnosis of streptococcal sore throat generally are unavailable, and empirical therapy is the rule—often self-treatment with a variety of suboptimal regimens.

OCR for page 338
New Vaccine Development: Establishing Priorities, Volume II, Diseases of Importance in Developing Countries The second preventive strategy, called “secondary prophylaxis,” consists of the continuous administration of antimicrobial agents to patients who have experienced an attack of ARF. This intervention is the most effective yet devised to prevent rheumatic fever and rheumatic heart disease. The approach is highly effective because it is directed at a group of patients at very high risk of developing repetitive episodes of ARF and of sustaining chronic valve damage. Even this therapy, however, has problems. As with primary prevention, secondary prevention is most difficult to accomplish in those areas where it is needed most. Identifying persons at risk, keeping them under continuous supervision, and arranging for monthly injections of benzathine penicillin G (the only practical means of prophylaxis in most developing countries) require a highly developed public health infrastructure, which is all too often lacking. Once ARF has developed, there is no known therapeutic intervention that can alter the risk of developing carditis. PROSPECTS FOR VACCINE DEVELOPMENT For many years, attempts have been made to develop safe and effective vaccines against strains of group A streptococci that give rise to ARF and rheumatic heart disease. Vaccination efforts have been hampered by toxic reactions to almost any streptococcal product injected into human subjects (Stollerman, 1975). Moreover, a number of vaccine preparations have been associated with antigens that evoke antibodies that cross-react with host tissues, especially heart sarcolemmal membranes (Kaplan, 1963). Protective immunity against group A streptococci is directed exclusively against the type-specific M protein on the surface of virulent organisms (Lancefield, 1962). Since this protein is often tenaciously associated with antigens that evoke toxic reactions, or worse, immunological cross-reactions with host tissues, efforts have been directed toward the isolation and purification of the protective M protein antigens. Recent studies of the immunogenicity of polypeptide fragments of M protein liberated from whole type 24 group A streptococci by limited peptic digestion have been promising. The purified extracts lacked detectable heart cross-reactive antigens, but retained protective determinants as demonstrated by vaccination of laboratory animals (Beachey et al., 1974, 1977) and by preliminary vaccine trials in human volunteers (Beachey et al., 1979). Animal trials with another serotype of M protein (pep M5) extracted with pepsin from “rheumatogenic” type 5 GrAS, however, resulted in the development of heart cross-reactive antibodies in one of the first three rabbits immunized with the purified protein (Dale and Beachey, 1982). Detailed analyses revealed that the cross-reactive determinant resided in the M protein molecule itself rather than in a contaminant. Furthermore, antibody directed toward the heart cross-reactive determinant was shown to be opsonic and protective against types 5 and 19 streptococci, both of which had been shown by Kaplan (1963) to be associated with type-specific heart cross-reactive antigens. Cross-

OCR for page 338
New Vaccine Development: Establishing Priorities, Volume II, Diseases of Importance in Developing Countries absorption studies showed that the cross-protective, heart cross-reactive antigen represented only a minor determinant of the type 5 M protein molecule, since other protective determinants (probably the majority) on the pep M5 molecule were clearly not heart cross-reactive (Dale and Beachey, 1982). This raised the hope that one might be able to cleave M protein and eliminate potentially harmful regions of the molecule while retaining protective regions. Immunogenicity of Native Peptide Fragments of M Protein Several lines of evidence support the notion evoking protective immune responses by immunization with peptide fragments of M protein. First, the intact M protein molecule appears to be considerably larger than the polypeptide extracted by most methods from whole streptococci or cell walls (Phillips et al., 1982); thus far, most M protein vaccines extracted by a number of different methods (Beachey et al., 1977; Fischetti et al., 1976; Fox and Wittner, 1969; Lancefield, 1962; Russell and Facklam, 1975; Vosti and Williams, 1978) and studied for immunogenicity have consisted only of large polypeptide fragments of the M protein molecule. Second, several laboratories have presented evidence that individual M protein molecules contain many antigenic determinants, most of which are type-specific (Beachey and Cunningham, 1973; Beachey et al., 1978; Cunningham and Beachey, 1975; Fischetti et al., 1976; Fox and Wittner, 1969; Hasty et al., 1982) but some of which are cross-reactive with other M serotypes (Beachey and Seyer, 1982; Dale et al., 1980; Fischetti 1977; Fox and Wittner, 1968; Wittner and Fox 1977). Third, purified subpeptides derived by cyanogen bromide cleavage of type 24 M protein each evoked protective antibodies against the homologous serotype of streptococci (Beachey et al., 1980). Finally, two chemically synthesized, 35-residue peptide fragments of pep M24 were shown to evoke similar type-specific protective antibodies without eliciting heart-reactive antibodies (Beachey et al., 1981a, 1983). Structure of M Protein Recent studies of the primary structure of three different serotypes of M protein have revealed several remarkable features. First, although the amino acid sequences of each of three different M serotypes (types 5, 6, and 24) are unique, certain amino acids are conserved among all three (Dale et al., 1980; Fischetti and Manjula, 1982; Manjula and Fischetti, 1980a, 1980b). Second, the M protein molecules are composed to varying degrees of internally repeating covalent structures (Beachey et al., 1978; Manjula and Fischetti, 1980b; Manjula et al., 1984). Third, the M proteins contain a seven-residue periodicity in their amino acid sequences that is reminiscent of an α-tropomyosin of muscle tissue (Hosein et al., 1979; Manjula and Fischetti, 1980b).

OCR for page 338
New Vaccine Development: Establishing Priorities, Volume II, Diseases of Importance in Developing Countries This periodicity accounts for the α-helical coiled-coil structure of the M protein molecule (Phillips et al., 1982); it is the first such conformation to be demonstrated for any surface appendage of bacteria (Phillips et al., 1982). Electron microscopy of the rotary shadowed coiled-coils of pepsin-, detergent-, and phage lysin-extracted M protein of type 6 streptococci suggests that the pepsin- and detergent-extracted M proteins represent the distal ends of these surface fibrillar structures, whereas the lysin-extracted M protein represents most, if not all, of these structures. Thus far, most of the primary structural studies have used pepsin or detergent extracts of M protein. Chemically Synthesized M Protein Fragments The synthesis of M protein polypeptides has been based on the covalent structures determined for the cyanogen bromide peptides of type 24 M protein (pep M24) and of the NH2-terminal sequence analysis of type 5 M protein. The synthetic peptides studied thus far include synthetic copies of CB3 and CB7, designated S-CB3 and S-CB7, and subpeptides of the 35-residue S-CB7, including S-CB7(13–35), S-CB7(18–35), S-CB7(23–35)Cys, and S-CB7(18–29). S-CB7(18–29) overlaps two subpeptides derived by digesting lysyl-blocked CB7 (see below). A 20-residue peptide of pep M5, S-M5(1–20) also has been synthesized. Each of the peptides was synthesized by the solid phase method of Merrifield (1963), as previously described (Beachey et al., 1981a, 1981b, 1983; Jolivet et al., 1984). The identity of the synthetic peptides has been confirmed by amino acid analyses and automated Edman degradation (Beachey et al., 1980). When emulsified in complete Freund’s adjuvant and injected as a single 25 nmol dose, S-CB7 evoked antibody titers at 6 weeks of 1:400, 1:1,280, and 1:25,600, respectively, in each of three rabbits, as determined by enzyme-linked immunosorbent assays (ELISA). Only the serum of the rabbit showing the highest ELISA titer was able to opsonize type 24 streptococci. In contrast, the same dose of S-CB7 covalently linked to polylysine and emulsified in complete Freund’s adjuvant evoked strong ELISA as well as opsonic antibody titers in each of three rabbits (Beachey et al., 1981a). The results obtained using S-CB3 instead of S-CB7 were virtually identical (Beachey et al., 1983). In opsono-bactericidal assays using types 5, 6, and 24 streptococci, the immune sera were capable of promoting phagocytosis of and killing only the type 24 streptococci, indicating that the humoral immune responses to S-CB7 were type-specific (Beachey et al., 1980). Immunodiffusion tests of the immune sera revealed precipitin arcs of identity between wells containing polylysine conjugates of native and synthetic CB7, as well as with the intact pep M24. There was no cross-reactivity of the anti-S-CB7 antiserum with pep M5 or pep M6. None of the S-CB7 immune sera reacted with the sarcolemmal membranes of frozen sections of human heart, as determined by immunofluorescence (Beachey et al., 1979). The protective activity of the S-CB7 immune sera was tested by positive mouse protection tests. Mice were injected with 0.2 m1 of a

OCR for page 338
New Vaccine Development: Establishing Priorities, Volume II, Diseases of Importance in Developing Countries pool of the immune sera and 24 hours later challenged with type 24 or type 6 streptococci. The immune sera clearly protected against the homologous type 24, but not against the heterologous type 6 streptococci (Beachey et al., 1981a). Having demonstrated that a protective determinant of M protein resides in a synthetic peptide as small as 35 residues long, it was of interest to determine whether protective determinants were retained in even smaller peptide fragments. CB7 was cleaved with trypsin after blocking lysine residues with maleic anhydride. Because CB7 contains only one arginine residue at position 23, the molecule was cleaved into a COOH-terminal dodecapeptide and an NH2-terminal 23-residue peptide. After deblocking the lysine residues with pyridine/acetate, the subpeptides were separated and purified by HPLC reverse-phase chromatography (Beachey et al., 1981a). The purified peptides were then tested for their abilities to inhibit opsonization by anti-S-CB7. The synthetic overlapping peptide, S-CB7(18–29), also was tested for inhibitory activity. The results clearly showed that the NH2-terminal 23-residue peptide and the COOH-terminal 12-residue peptide were inhibitory, although higher concentrations than that of the intact CB7 were required to achieve equal levels of inhibition. The overlapping 12-residue peptide, S-CB7(18–29), in contrast, was without effect even at concentrations as high as 100 nmol (Beachey et al., 1981a). These results indicate that a protective determinant of type 24 M protein resides in a peptide fragment containing only 12 residues. The inability of the overlapping synthetic peptide to block opsonization suggests that CB7 contains at least two distinct protective antigenic determinants, neither one of which resides in the overlapping peptide. That protective epitopes reside in the subpeptides of CB7 was confirmed by studies of the immunogenicity of the synthetic subpeptides S-CB7-(13–35), -(18–35), and -(23–35)C covalently linked to tetanus toxoid (Beachey et al., 1981b). Each of these peptides evoked opsonic antibodies in rabbits. The substitution, however, of a single amino acid in the 13-residue M protein peptide, S-CB7(23–35)Cys, at position 33 (lysine substituted for glycine) rendered this protective peptide nonprotective (Dale et al., 1980). In contrast to the high degree of type-specificity of the humoral immune responses to each of the synthetic peptides, the cellular immune responses were highly cross-reactive. The lymphocytes of rabbits immunized with S-CB7 were equally responsive in mitogenic assays to heterologous pep M5 as they were to the homologous pep M24 (Beachey et al., 1981a). Furthermore, immunization with S-CB7(18–29) covalently conjugated with polylysine, although failing to evoke effective humoral immunity, nevertheless induced non-type-specific cellular immunity similar to that of S-CB7. Neither of the synthetic peptides were of sufficient size to induce mitogenesis of lymphocytes from immunized rabbits (Beachey et al., 1980). These studies provided evidence in support of the hypothesis that limited peptide regions of the M protein molecule would suffice as protective immunogens, especially when covalently linked to carrier molecules, such as polylysine or tetanus toxoid. Furthermore, they suggested that one might select for synthesis small peptide regions of

OCR for page 338
New Vaccine Development: Establishing Priorities, Volume II, Diseases of Importance in Developing Countries M protein that contain protective, but not tissue cross-reactive, antigenic determinants. This led to an examination of the protective effects and heart cross-reactivity of a 20-residue synthetic peptide of type 5 M protein, a molecule demonstrated to contain within its covalent structure an epitope(s) that raises antibodies cross-reactive with a protein in the sarcolemma of human heart tissue. When covalently linked to tetanus toxoid with glutaraldehyde, the synthetic peptide S-M5(1–20) evoked antibodies that were protective against the related type 5 streptococci (Dale et al., 1983). Even though each of the animals received five times the dose, on a molar basis, required to produce heart cross-reactive antibody with the intact pep M5 molecule, none of the immunized animals developed heart cross-reactive antibodies. These studies should encourage the development of streptococcal vaccines composed of small synthetic peptide regions of M protein that contain protective, but not tissue cross-reactive, antigenic determinants. Knowledge of the complete covalent structures of several serotypes of M protein also should allow application of the principles recently reported by Hopp and Woods (1981) that predict which antigenic regions to isolate or synthesize based on the relative hydrophilicity of various peptide regions as determined by their amino acid sequences. At the time of publication, phase 1 studies in humans on tetanus toxoid conjugated M protein peptides were expected to begin shortly and plans for more extensive efficacy trials, possibly to be conducted in India and Chile, were being prepared (National Institute of Allergy and Infectious Diseases, 1985). Conclusion The whole M protein molecule is not needed to effectively immunize against group A streptococcal infections. Chemically synthesized peptide fragments of carefully chosen regions of various M protein molecules should provide a vaccine that can be administered safely to humans without fear of evoking host tissue cross-reactive antibodies. There is some evidence to suggest that certain protective determinants of M protein may be shared among different serotypes. The identification and synthesis of such common protective determinants should provide the basis for the formulation of a broadly protective vaccine against many serotypes of rheumatogenic group A streptococci. Extensive work is currently under way directed at improving the immunogenicity of peptides through coupling to carriers or by the use of adjuvants. This should enhance the prospects for the success of the approach discussed above. REFERENCES Beachey, E.H., and M.W.Cunningham. 1973. Type-specific inhibition of preopsonization versus immunoprecipitation by streptococcal M proteins. Infect. Immun. 8:19–24.

OCR for page 338
New Vaccine Development: Establishing Priorities, Volume II, Diseases of Importance in Developing Countries Beachey, E.H., and J.M.Seyer. 1982. Pp. 401–410 in Bacterial Vaccines, J.B.Robbins, J.C.Hill, and J.C.Sadoff, eds. New York: Thieme-Stratton. Beachey, E.H., G.L.Campbell, and I. Ofek. 1974. Peptic digestion of streptococcal M protein. II. Extraction of M antigen from group A streptococci with pepsin. Infect. Immun. 9:891–896. Beachey, E.H., G.H.Stollerman, E.Y.Chiang, T.M.Chiang, J.M.Seyer, and A.H.Kang. 1977. Purification and properties of M protein extracted from group A streptococci with pepsin: Covalent structure of the amino terminal region of type 24 M antigen. J. Exp. Med. 145(6):1479–1483. Beachey, E.H., J.W.Seyer, and A.H.Kang. 1978. Repeating covalent structure of streptococcal M protein. Proc. Natl. Acad. Sci. USA 75(7):3163–3167. Beachey, E.H., G.H.Stollerman, R.H.Johnson, I. Ofek, and A.L.Bisno. 1979. Human immune response to immunization with a structurally defined polypeptide fragment of streptococcal M protein. J. Exp. Med. 150(4):862–877. Beachey, E.H., J.M.Seyer, and A.H.Kang. 1980. Primary structure of protective antigens of type 23 streptococcal M protein. J. Biol. Chem. 255(13):6284–6289. Beachey, E.H., J.M.Seyer, J.B.Dale, W.A.Simpson, and A.H.Kang. 1981a. Type-specific protective immunity evoked by synthetic peptide of streptococcus pyogenes M protein. Nature 292(5822):457–459. Beachey, E.H., A.Tartar, J.M.Seyer, and L.Chedid. 1981b. Epitope-specific protective immunogenicity and chemically synthesized 13–, 18–, and 23-residue peptide fragments of streptococcal M protein. Proc. Natl. Acad. Sci. USA 81:2203–2207. Beachey, E.H., J.M.Seyer, J.B.Dale, and D.L.Hasty. 1983. Repeating covalent structure and protective immunogenicity of native and synthetic polypeptide fragments of type 24 streptococcal M protein. Mapping of protective and nonprotective epitopes with monoclonal antibodies. J. Biol. Chem. 258(21):13250–13257. Bisno, A.L. 1980. P. 789 in Streptococcal Diseases and the Immune Response, Streptococcal Disease and the World Status. New York: Academic. Bisno, A.L. 1982. Rheumatic fever. Pp. 1450–1457 in Cecil Textbook of Medicine, 16th ed., J.B.Wyngaarden and L.H.Smith, Jr., eds. Philadelphia: W.B.Saunders. Bisno, A.L. 1985. Classification of streptococci. Pp. 1123–1142 in Principles and Practice of Infectious Diseases, 2d ed., G.L. Mandell, R.G.Douglas, Jr., and J.E.Bennett, eds. New York: Wiley and Sons. Bisno, A.L., I.A.Pearce, and H.P.Hall. 1970. Contrasting epidemiology of acute rheumatic fever and acute glomerulonephritis. 1970. N. Engl. J. Med. 283:561–565. Brownell, R.D., and F.Bailen-Rose. 1973. Acute rheumatic fever in children. Incidence in a borough of New York City. JAMA 224:1593–1597.

OCR for page 338
New Vaccine Development: Establishing Priorities, Volume II, Diseases of Importance in Developing Countries Cunningham, M.W., and E.H.Beachey. 1975. Immunochemical properties of streptococcal M protein purified by isoelectric focusing. J. Immunol. 115:1002–1006. Dale, J.B., and E.H.Beachey. 1982. Protective antigenic determinant of streptococcal M protein stored with sarcolemmal membrane protein of human heart. J. Exp. Med. 156(4):1165–1176. Dale, J.B., I. Ofek, and E.H.Beachey. 1980. Heterogeneity of type-specific and cross-reactive antigenic determinants within a single M protein of group A streptococci. J. Exp. Med. 151(5):1026–1038. Dale, J.B., J.M.Seyer, and E.H.Beachey. 1983. Type-specific immunogenicity of a chemically synthesized peptide fragment of type 5 streptococcal M protein. J. Exp. Med. 158(5):1727–1732. Dodu, S. 1985. Personal communication, World Health Organization, Geneva, Switzerland. Fischetti, V.A. 1977. Streptococcal M protein extracted by nonionic detergent. II. Analysis of the antibody response to the multiple antigenic determinants of the M-protein molecule. J. Exp. Med. 146(4):1108–1123. Fischetti, V.A., and B.N.Manjula. 1982. Pp. 411–418 in Bacterial Vaccines, J.B.Robbins, J.C.Hill, and J.C.Sadoff, eds. New York: Thieme-Stratton. Fischetti, V.A., E.C.Gotschlich, G.Siviglia, and J.B.Zabriskie. 1976. Streptococcal M protein extracted by nonionic detergent. I. Properties of the antiphagocytic and type-specific molecules. J. Exp. Med. 144(1):32–53. Fox, E.N., and M.K.Wittner. 1968. Antigenicity of the M proteins of group A hemolytic streptococci. IV. Cross-reactivity between serotypes. J. Immunol. 100:39–45. Fox, E.N., and M.K.Wittner. 1969. New observations on the structure and antigenicity of the M proteins of the group A streptococcus. Immunochem. 6:11–24. Gordis, L., A.Lilienfeld, and R.Rodriguez. 1969. Studies in the epidemiology and preventability of rheumatic fever. I. Demographic factors and the incidence of acute attacks. J.Chron. Dis. 21:645–654. Hasty, D.L., E.H.Beachey, W.A.Simpson, and J.B.Dale. 1982. Hybridoma antibodies against protective and nonprotective antigenic determinants of a structurally defined polypeptide fragment of streptococcal M protein. J. Exp. Med. 155(4):1010–1018. Hopp, T.P., and K.R.Woods. 1981. Prediction of protein antigenic determinants from amino acid sequences. Proc. Natl. Acad. Sci. USA 78(6):3824–3828. Hosein, B., M.McCarty, and V.A.Fischetti. 1979. Amino acid sequence and physiochemical similarities between streptococcal M protein and mammalian tropomysin. Proc. Natl. Acad. Sci. USA 76(8):3765–3768. Jolivet, M., F.Audibert, E.H.Beachey, A.Tartar, H. Gras-Masse, and L.Chedid. 1984. Epitype specific immunity elicited by a synthetic streptococcal antigen without carrier or adjuvant. Biochem. Biophys. Res. Commun. 117(2):359–366.

OCR for page 338
New Vaccine Development: Establishing Priorities, Volume II, Diseases of Importance in Developing Countries Kaplan, M.H. 1963. Immunologic relation of streptococcal and tissue antigens. I. Properties of an antigen in certain strains of group A streptococci exhibiting an immunologic cross-reaction with human heart tissue. J. Immunol. 90:595–606. Lancefield, R.C. 1962. Current knowledge of type-specific M antigens of group A streptococcus. J. Immunol. 89:307–313. Lancet. 1985. Decline in rheumatic fever. Lancet II:647–648. Land, M.A., and A.L.Bisno. 1983. Acute rheumatic fever. A vanishing disease in suburbia. JAMA 249(7):895–898. Manjula, B.N., and V.A.Fischetti. 1980a. Studies on group A streptococcal M-proteins: Purification of type 5 M-protein and comparison of its amino terminal sequence with two immunogenically unrelated M-protein molecules. J. Immunol. 124(1):261–267. Manjula, B.N., and V.A.Fischetti. 1980b. Tropomyosin-like seven residue periodicity in three immunogenically distinct streptococcal M proteins and its implication for the antiphagocytic property of the molecule. J. Exp. Med. 151(3):695–707. Manjula, B.N., A.S.Acharya, S.M.Mische, T.Fairwell, and V.A. Fischetti. 1984. The complete amino acid sequence of a biologically active 197-residue fragment of M protein isolated from type 5 group A streptococci. J. Biol. Chem. 259(6):3686–3693. Markowitz, M. 1981. Observations on the epidemiology and preventability of rheumatic fever in developing countries. Clin. Ther. 4:240–251. Merrifield, R.B. 1963. Solid phase peptide synthesis. I. The synthesis of a tetrapeptide. J. Am. Chem. Soc. 85:2149–2155. National Institute of Allergy and Infectious Diseases. 1985. Program on Accelerated Development of New Vaccines. Progress Report. Bethesda, Md.: National Institutes of Health. Phillips, G.N., P.F.Flicker, C.Cohen, B.N.Manjula, and V.A. Fischetti. 1982. Streptococcal M protein: Alpha-helical coiled-coil structure and arrangement on the cell surface. Proc. Natl. Acad. Sci. USA 78(8):4689–4693. Quinn, R.W. 1982. Streptococcal infections. Pp. 525–552 in Bacterial Infections of Human, A.S.Evans and H.A.Feldman, eds. New York: Plenum. Quinn, R.W., and C.F.Federspiel. 1974. The incidence of rheumatic fever in metropolitan Nashville, 1963–1969. Am. J. Epidemiol. 99:273–280. Russell, H., and R.R.Facklam. 1975. Guanidine extraction of streptococcal M proteins. Infect. Immun. 12(3):679–686. Stollerman, G.H. 1975. Rheumatic Fever and Streptococcal Infection. New York: Grune and Stratton. Vosti, K.L., and W.K.Williams. 1978. Extraction of streptococcal type 12 M protein by cyanogen bromide. Infect. Immun. 21(2):546–555. Whitnack, E., and A.L.Bisno. 1980. Rheumatic fever and other immunologically mediated cardiac diseases. Pp. 894–929 in Clinical Immunology, Vol. 2, C.W.Parker, ed. Philadelphia: W.B.Saunders.

OCR for page 338
New Vaccine Development: Establishing Priorities, Volume II, Diseases of Importance in Developing Countries Wittner, M.K., and E.N.Fox. 1977. Homologous and heterologous protection of mice with group A streptococcal M protein vaccine. Infect. Immun. 15(1):104–108. World Health Organization. 1980. Community control of rheumatic heart disease in developing countries. 1. A major public health problem. WHO Chron. 34:336–345. United Kingdom and United States Joint Report. 1965. Circ. 32:457.