Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Pneumococcal Conjugate Vaccine and Pneumococcal Common Protein Vaccines
The original demonstration by Avery and Goebel, in 1929, that capsular polysaccharides were immunogenic when linked to a protein, and 60 years later, the demonstrated efficacy of Haemophilus influenzae type b (Hib) conjugate vaccines in the 1990s (see Chapter 25) led to an extraordinary public health intervention in the year 2000 when conjugated pneumococcal polysaccharide vaccines were introduced for the prevention of pneumococcal disease in infants. Acute respiratory infections are a major global cause of death in young children and in the elderly, but pneumonia deaths from 1990 to 2019 have decreased by 62.5% (95% CI, 54.9%–69%). While increased access to antibiotics likely played a major role, the introduction of both Hib and pneumococcal conjugate vaccines (PCVs) has also had a substantial impact.
This chapter reviews pneumococcal polysaccharide-protein conjugates, isolated pneumococcal proteins, and whole-cell pneumococcal vaccines as avenues for wider coverage and increased efficacy compared with pneumococcal polysaccharide vaccine (PPV) (reviewed in Chapter 48). Conjugate vaccines are the most advanced of these approaches with large-scale global rollout and randomized clinical trials in multiple countries confirming their efficacy against invasive disease and otitis media. Conjugate vaccines are widely available in most developed countries, with introduction into more than 60 of the 73 Global Alliance for Vaccines and Immunization (GAVI)-eligible developing countries by the end of 2020. The overall effectiveness of conjugate pneumococcal vaccines has resulted in large part from the impact on vaccine serotype-specific carriage and thus reduced transmission of vaccine types in the community. In the United States, immunization of young children has led to the virtual elimination of vaccine-type disease in both children and adults, , and the more than a 90% reduction in hospitalization for nonbacteremic pneumonia is attributable to herd protection of adults resulting from the vaccination of children. At the same time as protection is largely limited to the specific vaccine serotypes, vaccine effectiveness has been reduced by replacement carriage resulting from nonvaccine serotypes and replacement disease resulting from the disease potential of those serotypes. The impact of PCVs on herd protection in developing countries has also been seen. A randomized trial of PCV in the elderly demonstrated efficacy against both bacteremic and nonbacteremic pneumonia caused by vaccine types. The development of pneumococcal proteins or whole cell vaccines are yet to demonstrate a proof of concept in humans. A vaccine that combined conjugated pneumococcal polysaccharides with additional nonconjugated pneumococcal proteins failed to impact otitis media in a clinical trial. Studies of other vaccine prototypes are ongoing, as will be subsequently described.
PATHOGENESIS
With the possible exception of the great apes, humans are the only natural host of the pneumococcus. However, other mammals, such as pets (e.g., horses, dogs, cats, and guinea pigs), zoo animals (e.g., dolphins), and laboratory animals (e.g., mice and rats) can acquire the organism from their handlers. The organism lives in the nasopharynx of humans, from whence it can be transmitted by respiratory droplets to other individuals. In most cases, pneumococci are carried asymptomatically before being cleared from the nasopharynx. The duration of carriage is age- and serotype-dependent. In some cases, usually shortly after the acquisition of carriage, pneumococci can cause mucosal disease by local spread to the middle ear (causing otitis media), the sinuses (causing sinusitis), or the lungs (causing pneumonia). In addition, pneumococci may cause systemic infections with considerable risk of mortality, including bacteremia and meningitis and less commonly infections in joints, bones, and soft tissues.
The pneumococcal capsular polysaccharide is the most important virulence factor for Streptococcus pneumoniae , , although not the only one. The polysaccharide capsule significantly increases the resistance of pneumococci to phagocytosis by several mechanisms, including shielding from antibodies and complement. The capsule promotes adherence to epithelial cells in animal models. Strains of pneumococci that lack a detectable capsule can cause conjunctivitis, but are infrequently isolated in cases of invasive disease. Pneumococcal virulence depends in part on several serotype-specific characteristics, such as the chemical composition and molecular size of the capsular polysaccharide. , The different pneumococcal serotypes vary in virulence, depending on their ability to activate the alternative and classical complement pathway, to deposit and degrade the complement components on the capsule, and to resist phagocytosis , ; they also differ in their ability to induce antibodies. Antibodies to the capsular polysaccharides are protective, as has been amply demonstrated in animal models and in clinical trials of conjugate vaccines. ,
BACKGROUND
The reader is referred to Chapter 48 for a discussion of the clinical and bacteriologic aspects of pneumococcal disease.
EPIDEMIOLOGY
Pneumococcal Infections in Children Prior to Conjugate Vaccines
Before the introduction of seven-valent PCV7 in the United States, S. pneumoniae caused approximately 17,000 cases of invasive disease per year in children younger than 5 years of age, resulting in 700 cases of meningitis and 200 deaths. The annual incidence of invasive pneumococcal diseases was highest in children 6–11 months of age, at 235 per 100,000, declined to 35.2 per 100,000 in children between 2 and 4 years of age, and reached its lowest levels in children 5–17 years of age (3.9 per 100,000). A similar distribution of incidence by age was seen in Canada, although the rates for the youngest children were somewhat lower, peaking at 161.2 per 100,000 for children 6–17 months of age.
In Europe, the highest levels of invasive pneumococcal disease were in children in the second 6 months of life, then in those 2 years of age or younger, and declining steadily thereafter into the teenage years. , However, the measured incidence of invasive pneumococcal infections has generally been much lower in Europe than in the United States. For example, in England and Wales, the annual incidence of invasive pneumococcal diseases in infants between 6 and 11 months of age was 35.8 per 100,000 in the late 1990s, compared with 235 per 100,000 in the United States. In Finland, the annual incidence of invasive pneumococcal diseases was 45.3 per 100,000 among those younger than 2 years of age. This difference in incidence was thought to be largely a consequence of the common practice of obtaining blood cultures in outpatients in the United States, whereas in Europe they generally only obtained blood cultures with hospitalization.
Bacteremia without focus is more common in young children than in older persons, , , , and this diagnosis comprises 60–70% of invasive pneumococcal disease diagnosed in young children in the United States. , , Pneumonia and meningitis are the next most common, but meningitis remains the diagnosis with the highest case fatality rate. , , After the introduction of the Hib vaccine causing a marked decline in Hib disease, S. pneumoniae became the leading cause of bacterial meningitis and occult bacteremia in the United States. , As children grow older, the proportions of pneumococcal infections classified as meningitis and bacteremia without focus decline, and the proportion of pneumonia increases. , , ,
In the United States before the introduction of PCV, there were approximately 15 million office visits annually for the diagnosis of acute otitis media (AOM), with costs estimated at US$5 billion per year. S. pneumoniae was the most commonly reported bacterial cause of AOM, with the proportion of the pneumococcus as an etiologic agent recovered from middle ear fluid by tympanocentesis varying from 18% to 55% in published studies.
The burden of pneumococcal disease has been greatest for those populations in which the plain PPV has been the least immunogenic: children younger than 2 years of age; the elderly (≥65 years); immunocompromised individuals, including those infected with HIV; and those with malignancies. There is also a significant burden in individuals with functional or anatomic asplenia and those with chronic cardiovascular, pulmonary, or liver diseases. , Among individuals without underlying illness, children younger than age 2 years are most vulnerable to pneumococcal infections. There is evidence that this level of risk is increased further by spending time in crowded or communal settings, such as daycare centers. ,
Children from specific minority groups suffer disproportionately. Although the overall incidence of invasive pneumococcal disease in children younger than 2 years of age in the United States before licensure of PCV7 was 167 cases per 100,000 person-years, black children of the same age had an incidence of 400 per 100,000 person-years, Alaskan Native children had an incidence of 624 per 100,000 person-years, and Native American children had an overwhelming incidence of 2396 cases per 100,000 person-years ( Table 47.1 ), although rates in Navajo children did decline somewhat from those levels pre-PCV7. Asians and Hispanics also had a greater risk of pneumococcal pneumonia pre-PCV7 than did whites. Similar to the United States, indigenous people in Australia had much higher rates of pneumococcal disease with indigenous children younger than 2 years of age having an incidence of invasive pneumococcal diseases of 2053 cases per 100,000 person-years. Nonindigenous children in Australia had rates intermediate between those of Europe and the United States. Minority populations in Israel, including the Bedouins, also had a higher incidence of pneumococcal infections, although the difference was not as pronounced as in Australia and the United States. This difference may be socioeconomically based because minority groups in Israel are not a single ethnic population but, as a group, have higher birthrates, more crowded living conditions, less education, and lower socioeconomic status than the majority Jewish population. ,
TABLE 47.1
Incidence of Invasive Pneumococcal Disease in Children in the United States Before the Introduction of Pneumococcal Conjugate Vaccine by Age and Ethnic Group
| Age (years) | Incidence per 100,000 Person-Years | |||
|---|---|---|---|---|
| All Races | Blacks | Alaskan Natives | Native American (Navajo or Apache) | |
| <2 | 167 | 400 | 624 | 2,396 |
| 2–4 | 36 | 116 a | 98 b | 227 b |
| 5–9 | 6 | 9 | 23 | 54 |
| 10–19 | 3 | 5 | 5 | 35 |
Modified from Advisory Committee on Immunization Practices. Preventing pneumococcal disease among infants and young children. Recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Morb Mortal Recomm Rep . 2000;49(RR-9):1–35 .
Boys are more likely than girls to contract pneumococcal infections, , and there are often seasonal peaks in incidence with infection more likely in the winter. , Documented influenza or another respiratory viral infections are important risk factors for severe pneumococcal pneumonia and invasive disease. Recent ear infections, indicating the middle ear is an important route of S. pneumoniae invasion, and a lack of breastfeeding are also associated with increased incidence of invasive pneumococcal diseases.
Children with cochlear implants have an increased risk of pneumococcal meningitis, approximately 40 times higher than the rate in the general population of children. The risk was highest among children using a specific spacer, which was subsequently discontinued, and in those with middle ear malformations associated with cerebrospinal fluid leaks. A study in Denmark suggested that deaf children without cochlear implants have a fivefold increased risk of meningitis relative to normal hearing children.
Most deaths among children from pneumococcal disease occur in developing countries. In those areas, carriage of pneumococci frequently occurs before 2 months of age (in contrast, children in the United States acquire their first strain on average at 6 months of age), a larger proportion of invasive diseases occur by 6 months of age than in industrialized countries, and neonatal infections are not uncommon.
A compromised immune system, whether caused by HIV infection, malignancy, asplenia, or primary immunodeficiencies, is a major risk factor for pneumococcal disease. Infection with S. pneumoniae occurs more than 100 times more frequently in HIV-infected children than among the general population. Meningitis or recurrent bacterial sepsis can be the first indicator of AIDS in pediatric patients, , and otitis media is reported three times more frequently in HIV-infected children than in uninfected children. In the United States, rates of invasive pneumococcal diseases for HIV-infected children younger than 3 years of age are 12.6 times higher than in HIV-negative children, and in children younger than 5 years of age are 2.8 times higher than in HIV-negative children. In both children and adults, the introduction of highly active antiretroviral therapy (HAART) reduced the incidence of pneumococcal infections in HIV-infected patients, but rates remain elevated compared with HIV-uninfected persons. ,
In South Africa, the incidence of pneumococcal bacteremia was 36.9-fold higher in HIV-infected children than in HIV-negative children before introduction of antiretrovirals and PCV. The rates of pneumococcal bacteremia doubled between 1987 and 1997 as the rates of HIV infection rose. Overall, the disease spectrum is similar in HIV-positive and HIV-negative populations, except that pneumonia is more common in HIV-infected children.
Individuals with sickle cell disease and other hemoglobinopathies resulting in functional asplenia are also at increased risk for serious infections with S. pneumoniae . Without immunization, the rate of invasive pneumococcal disease in children with sickle cell disease is 30–100 times greater than that of healthy children of similar age. Even with the use of PCV and PPV, invasive pneumococcal disease among children with sickle cell disease who are younger than 5 years of age is 10-fold more common than in children without sickle cell disease, , and the case-fatality rate is much higher in children with sickle cell disease or other forms of anatomic or functional asplenia than in children with normal splenic function.
Children with nephrotic syndrome also are more susceptible to pneumococcal disease, with most infections occurring within the first 2 years after presentation. S. pneumoniae is responsible for 50–60% of sepsis and peritonitis in children with nephrotic syndrome. Patients who have received a bone marrow transplant are also more vulnerable to pneumococcal disease than the general population, in part because they often have impaired opsonization function and lose protection from previous immunizations because of T-cell and B-cell depletion. In addition, solid-organ transplantation also confers a higher risk of invasive pneumococcal disease. Of interest, the increased risk of pneumococcal infection in patients with Hodgkin disease can persist long after the underlying condition has been successfully treated.
Serotype Distribution
Forty serogroups comprising 100 serotypes of S. pneumoniae have been described. Among this large number of serotypes, a relatively small number account for most invasive disease in young children globally ( Fig. 47.1 ). Candidate pneumococcal conjugate vaccines have been designed based on the relative percentages of disease-causing serotypes. The first to be licensed in 2000 was a seven-valent vaccine (PCV7), sold under the trade name Prev(e)nar by Pfizer. The 10-valent vaccine (PCV10, Synflorix, sold by GlaxoSmithKline) is licensed in many other countries, but not in the United States. The 13-valent (PCV13, Prevenar 13, Pfizer) was licensed in the United States in 2010 and replaced PCV7. In 2020, the first conjugate pneumococcal vaccine produced by a developing country manufacturer (Serum Institute of India in Pune, India) was licensed and WHO prequalified, also comprising 10 serotypes but different from the GlaxoSmithKline product in replacing serotypes 4 and 18C with 6A and 19A. As of 2021, there are two next generation PCV’s licensed in the USA in adults only, PCV15 (adding serotypes 22F and 33F to PCV13) from Merck, and PCV20 from Pfizer adding serotypes 8, 10A, 11A, 12F, 15B, 22F, and 33F to PCV13. Although these next generation vaccines were licensed in the United States for adults in 2021, they are not expected to be licensed in children before 2023. Table 47.2 summarizes the serotypes included in each of these vaccines.
TABLE 47.2
Medical Conditions or Other Indications for Administration of PCV13 a and Indications for PPSV23 b Administration and Revaccination for Children Aged 2–18 Years c
| Risk Group | Underlying Medical Condition | PCV13 | PPSV23 | |
|---|---|---|---|---|
| Recommended | Recommended | Revaccination 5 y After First Dose | ||
| Immunocompetent persons | Chronic heart disease d | X | X | |
| Chronic lung disease e | X | X | ||
| Diabetes mellitus | X | X | ||
| Cerebrospinal fluid leaks | X | X | ||
| Cochlear implants | X | X | ||
| Alcoholism | X | X | ||
| Chronic liver disease | X | X | ||
| Cigarette smoking | X | X | ||
| Persons with functional or anatomic asplenia | Sickle cell disease/other hemoglobinopathies | X | X | X |
| Congenital or acquired asplenia | X | X | X | |
| Immunocompromised persons | Congenital or acquired immunodeficiencies f | X | X | X |
| Human immunodeficiency virus infection | X | X | X | |
| Chronic renal failure | X | X | X | |
| Nephrotic syndrome | X | X | X | |
| Leukemia | X | X | X | |
| Lymphoma | X | X | X | |
| Hodgkin disease | X | X | X | |
| Generalized malignancy | X | X | X | |
| Iatrogenic immunosuppression g | X | X | X | |
| Solid-organ transplantation | X | X | X | |
| Multiple myeloma | X | X | X | |
a 13-Valent pneumococcal conjugate vaccine.
b 23-Valent pneumococcal polysaccharide vaccine.
c Since 2010 the recommendation is that all healthy children younger than 5 years of age receive PCV13; PPV23 is not recommended for healthy children younger than 2 years of age.
d Including congestive heart failure and cardiomyopathies.
e Including chronic obstructive pulmonary disease, emphysema, and asthma.
f Including B-(humoral) or T-lymphocyte deficiency, complement deficiencies (particularly C1, C2, C3, and C4 deficiencies), and phagocytic disorders (excluding chronic granulomatous disease).
g Diseases requiring treatment with immunosuppressive drugs, including long-term systemic corticosteroids and radiation therapy. 715
Although comprehensive information about serotypes is not available for some regions, a systematic review summarized the information available from children younger than age 5 years before the introduction of pneumococcal conjugate vaccines. The review included 169 studies in the final analysis, representing 70 countries and 60,090 serotyped isolates. According to this analysis, serotypes 14, 6B, 1, 23F, 5, and 19F, in that order, were the most common serotypes found among sterile-site isolates globally. However, the distribution of disease-causing serotypes varied somewhat with location, age, and clinical presentation. Before the introduction of PCV7, serogroups in the seven-valent vaccine were responsible for more than 80% of invasive pneumococcal diseases in young children in the United States and Canada, and for 50% or more in all other regions; 10-valent and 13-valent vaccine formulations covered more than 70% of the invasive strains in all regions. Serotype 6A accounts for a substantial portion of disease, but within-serogroup cross-reactivity with serotype 6B in PCV7 and PCV10 provides protection.
Serotypes 1 and 5 are encountered much more frequently in the developing world than in the developed world, although between 2000 and 2010 serotype 1 disease increased in Western Europe and was found among children living in Utah. , Differences in blood culture practices may also affect the observed distribution of serotypes: virulent serotypes that are more likely to cause severe disease requiring hospitalization would have a higher apparent prevalence in regions that perform blood cultures solely in hospitalized patients. This may be the case for serogroups 1, 5, and 7, which were isolated at similar (low) rates in the United States and Europe, although the most common serotypes/groups (14, 6, 19, 18, 23, 9, and 4) had a much higher incidence in the United States before PCV use. For example, serotype 1 caused 5% of reported invasive pneumococcal diseases in Western Europe and only 0.5% of invasive pneumococcal diseases in the United States before conjugate vaccine introduction, yet the rate of serotype 1 disease was approximately 0.9 per 100,000 person-years in each region. This pattern indicates that serogroups 1, 5, and 7 cause severe rather than mild disease in both regions and supports the idea that certain serogroups are intrinsically highly virulent. This could also account, in part, for the high prevalence of serotypes 1 and 5 in the developing world, where, as in Europe, blood cultures are rarely performed on an outpatient basis.
Serotype distributions change with age. A narrower range of serotypes are responsible for disease in young children , than in older children and adults, and in most regions, 50% of invasive pneumococcal disease is caused by three serotypes in young children and by four or five serotypes in older children and adults. Serotypes 1, 5, and 18C are responsible for a larger portion of invasive pneumococcal disease in children. , ,
In addition to varying with age, certain serotypes are more likely to be associated with particular clinical presentations, although none of the most prominent serotypes is found exclusively in one clinical site, and all have the potential to invade multiple sites. Note that age and site of isolation may not be independent because, for example, young children are much more likely to present with meningitis than older children, and serotypes isolated from young children are therefore more likely to be found in the cerebrospinal fluid (CSF) than serotypes isolated from older children. Generally speaking, serogroups 6, 10, and 23 are more likely to be isolated from CSF, whereas serogroups 1, 4, and 14 are more often isolated from blood than from CSF. , Severe pneumonia is more likely to be associated with serotypes 1 and 3, although both pneumonia and these particular serotypes are more common in older children than the very young. Serotypes 19F, 23F, 14, and 6B are consistently among the six serotypes most frequently isolated from children with AOM, although their relative frequency varies by study. Serotype 3 is also a frequent AOM isolate in some areas, but not others. , ,
Antibiotic Resistance
Antibiotic resistance also varies with age, race, and geographic location. Globally, the northern European countries of Norway, Sweden, Denmark, The Netherlands, and Switzerland have consistently low rates of resistance; in contrast, France, Spain, Hong Kong, Singapore, and South Africa have high rates of resistance. In the United States, resistant strains are found more frequently in the southern states of Tennessee and Georgia than elsewhere. , The highest percentage of resistant strains occurs in children younger than 5 years of age with high rates of antibiotic use since the 1990s. , After 2000, increases in nonvaccine serotype 19A strains resistant to multiple antibiotics were seen in countries using seven-valent PCV7, such as the United States, , but also in countries that had not yet introduced PCV7. The amount and types of antibiotics used have been correlated with increased resistance. , , The serotypes most likely to become resistant (6A, 6B, 9V, 14, 19A, 19F, and 23F) , , were those that were frequently carried and most likely to cause AOM, and resistance rates are higher in isolates from middle ear fluid and sinuses than from normally sterile sites. This may be because pneumococci in these body sites are exposed to antibiotics more extensively than serotypes that cause invasive disease, both because they are carried for longer periods and because AOM is frequently treated with antibiotics. Most of the serotypes with the highest rates of resistance were included in the PCV7 vaccine or were within the same serogroups, and introduction of PCV7 reduced disease caused by resistant infections in the United States and in developing countries, both among children and among older persons through reduced transmission. , With the introduction on PCV13, serotype 19A, an often-resistant serotype that emerged after widespread PCV7 use, was targeted. , , ,
HISTORY OF VACCINE DEVELOPMENT AND VACCINE FORMULATIONS
Over the past 30 years, several PCV formulations have been tested in clinical trials with the vaccines differing in their included serotypes and carrier proteins (see Table 47.2 ). The number of pneumococcal serotypes included in current vaccines and in those in clinical development ranges from 10 to 20, still fewer than the 23 serotypes included in PPV. Although it would be preferable to include a larger number of different polysaccharides in a conjugate vaccine, doing so is technically challenging. Moreover, the total amount of carrier protein in the final vaccine may need to be limited because too much of some carrier proteins can impair the antibody response to the polysaccharide antigens. In developing countries, where the diversity of invasive disease causing serotypes in children is greater than in the USA, the incremental benefits in coverage for young children gained by increasing the number of serotypes beyond 13 may be greater.
Prevnar 13 (Pfizer), a pneumococcal 13-valent vaccine, is a sterile suspension of saccharides of the capsular antigens of S. pneumoniae serotypes 1, 3, 4, 5, 6A, 6B, 7F, 9V, 14, 18C, 19A, 19F, and 23F, individually linked to nontoxic diphtheria CRM 197 protein. Each serotype is grown in soy peptone broth. The individual polysaccharides are purified through centrifugation, precipitation, ultrafiltration, and column chromatography. The polysaccharides are chemically activated to make saccharides, which are directly conjugated by reductive amination to the protein carrier CRM 197 , to form the glycoconjugate. CRM 197 is a nontoxic variant of diphtheria toxin isolated from cultures of Corynebacterium diphtheriae strain C7 (β197) grown in a casamino acids and yeast extract-based medium. CRM 197 is purified through ultrafiltration, ammonium sulfate precipitation, and ion-exchange chromatography. The individual glycoconjugates are purified by ultrafiltration and column chromatography and analyzed for saccharide to protein ratios, molecular size, free saccharide, and free protein. The individual glycoconjugates are compounded to formulate Prevnar 13. Potency of the formulated vaccine is determined by quantification of each of the saccharide antigens and by the saccharide to protein ratios in the individual glycoconjugates. Each 0.5 mL dose of the vaccine is formulated to contain approximately 2.2 µg of each of S. pneumoniae serotypes except 4.4 µg of 6B saccharides, 34 µg CRM 197 carrier protein, 100 µg polysorbate 80, 295 µg succinate buffer, and 125 µg aluminum as aluminum phosphate adjuvant.
The first licensed PCV7 from Wyeth, also conjugated to CRM 197 , included 2 µg each of the polysaccharide serotypes. Several conjugate vaccines that were tested in clinical trials are no longer in development. Merck developed a seven-valent vaccine containing polysaccharides of the same serotypes as the Wyeth product, conjugated to a meningococcal outer membrane protein (OMP) modified from Neisseria meningitidis, and Wyeth produced a CRM 197 -conjugated nine-valent vaccine containing 2 µg each of polysaccharide from serotypes 1 and 5 in addition to the serotypes in the seven-valent vaccine. Sanofi Pasteur developed an 11-valent vaccine that added serotypes 3 and 7F; in this vaccine (PncD/T), some of the saccharides (serotypes 3, 6B, 14, and 18C) were conjugated to diphtheria toxoid and the others (serotypes 1, 4, 5, 7F, 9V, 19F, and 23F) were conjugated to tetanus protein. Other potential carrier proteins that have been tested in animal studies include bovine serum albumin, human immunoglobulin (Ig) G, complement C3d, keyhole limpet hemocyanin, flagellar protein of Salmonella , pertussis toxoid, and pneumolysin toxoid. The conjugation chemistry and manufacturing protocols vary for each preparation.
GlaxoSmithKline designed an 11-valent vaccine in which 1 µg of each saccharide was conjugated to protein D (9–16 µg per dose of vaccine) derived from nontypeable H. influenzae as a carrier protein for eight of the 10 serotypes (1, 4, 5, 6B, 7F, 9V, 14, and 23F), while serotypes 18C and 19F are conjugated to tetanus toxoid (5–10 µg) and to diphtheria toxoid (3–6 µg), respectively. All conjugates are adsorbed onto 0.5 mg aluminum phosphate. This vaccine protected against both pneumococcal and H. influenzae otitis media. However, serotype 3 proved not to be protective against AOM in the clinical trial ; consequently, GlaxoSmithKline proceeded with the clinical development of a 10-valent version, containing serotypes 1, 5, and 7F in addition to those in the seven-valent vaccine. This 10-valent version (PCV10, sold under the name Synflorix) is currently licensed in many countries globally and is prequalified by the World Health Organization (WHO). This vaccine is not licensed in the United States, but was used for a time in Quebec. Merck has an investigational 15-valent vaccine that has completed Phase III trials in adults and infants. This vaccine is also conjugated to CRM 197 , and in addition to the serotypes in the 13-valent vaccine, includes serotypes 22F and 33F. The vaccine contains 2 µg of each serotype, except for 4 µg of serotype 6B, and uses aluminum phosphate as an adjuvant.
The Serum Institute of India has a 10-valent CRM 197 conjugate PCV licensed in India and prequalified by WHO in 2020. The vaccine has been formulated to contain important serotypes seen in developing countries and includes serotypes 1, 3, 5, 6A, 6B, 7F, 14, 19A, 19F, and 23F. A 13-valent conjugate vaccine made by Walvax Biotechnology has recently been approved for market use in China.
Both 15 valent from Merck and 20 valent next generation vaccines have undergone clinical trials in adults and infants, although they are currently licensed only for adults. They are formulated like PCV13 but have the following 7 additional serotypes conjugated to CRM197; serotypes 8, 10A, 11A, 12F, 15B, 22F, and 33F. in the 20 valent and just 22F and 33 F additional in the 15 valent from Merck. A number of additional conjugate vaccines are in clinical development including those produced by Panacea and BioE in India, and SK Chemicals in South Korea. Preclinical products of higher valency than 20 have been announced by three US companies; Sutrovax, Inventprise and Affinivax, the latter having recently completed a Phase II clinical trial with a 24-valent pneumococcal vaccine in older adults. A seven-valent PCV that includes serotypes 1 and 5 has been tested in humans in Cuba, but has not yet been licensed widely or used outside Cuba.
IMMUNOGENICITY
Measurement of Immunogenicity
For many encapsulated bacterial pathogens, antibodies against their capsular polysaccharides are protective against disease. In pneumococcal infections, this has been demonstrated indirectly by the increased incidence of serious pneumococcal infections in individuals with lower antibody concentrations (e.g., infants and the elderly, patients with hypogammaglobulinemia or agammaglobulinemia, and patients recovering from bone marrow transplantation). Direct evidence of the protective effect of anticapsular antibodies comes from studies where passively administered antibodies have provided protection from otitis media or invasive pneumococcal disease. , Furthermore, vaccines that induce antibodies to pneumococcal capsular polysaccharides are protective against invasive pneumococcal infections (see Chapter 48).
The two most important serologic methods used to quantify and evaluate the function of antibodies to the capsule induced by vaccination are IgG quantification by enzyme immunoassay (EIA) and measurement of opsonophagocytic activity (OPA) in sera from vaccinated individuals. The EIA method has been validated extensively, a link between antibody concentration and efficacy against invasive pneumococcal disease has been established, , and a process for cross-laboratory standardization of the assay has been completed. The OPA assay provides important information about the functional activity of the antibodies and has also been the focus of intensive standardization efforts.
Additional antibody studies, such as isotype and subclass analysis, determination of antibody avidity, and serotype cross-reactivity, provide qualitative information about the immune mechanisms evoked by the vaccines. However, these additional assays have not been standardized, and their value in predicting protective efficacy has not been determined. Further information about the characteristics of the immune response, including induction of immunologic memory as well as cellular and mucosal immune responses to the vaccine is under investigation. Additional immunologic data may be obtained through challenge studies in animal models.
Enzyme Immunoassay
Immune responses to pneumococcal vaccines have been evaluated by radioimmunoassay and EIA, both of which measure the binding of antibodies to pneumococcal polysaccharide antigens. While the radioimmunoassay technique is no longer in use, EIA remains a sensitive, convenient, and rapid method for evaluating a large number of samples. Concentrations of IgA, IgM, and IgG isotypes, as well as of IgG subclasses, can be determined using a reference serum. The United States Food and Drug Administration (FDA) previously made available a reference serum from pooled human sera (89SF) to use to standardize the assay. More recently, a new serum pool was formulated by the FDA (007Sp) and validated by the WHO; it replaced 89SF as the pneumococcal reference standard.
All current EIA protocols have an absorption step to remove anti-C-polysaccharide antibodies, , the presence of which was the main limitation of the earlier radioimmunoassays. However, it has also been recognized that, to some extent, similar problems exist with EIA, in that some sera contain polyreactive antibodies that recognize pneumococcal polysaccharides of different serotypes. , , These antibodies are especially prevalent in the sera of individuals who have naturally acquired antibodies and are less prevalent in the sera of those immunized with pneumococcal vaccines. These antibodies are likely directed against the contaminants in the pneumococcal polysaccharide preparations that are used as antigens in the assays or directed to polyreactive epitopes present in the polysaccharide antigens. Inhibiting antibody binding using soluble polysaccharides of another serotype (e.g., of serotype 22F) removes these polyreactive antibodies, improves the specificity of the assay, and increases the linear correlation between anticapsular polysaccharide antibody concentration and OPA. , Consequently, neutralization with 22F is now included in the EIA protocol. This change has caused some confusion regarding the need to change threshold values for protection as discussed below. Further complicating the picture for the Merck 15-valent vaccine in development, which contains serotype 22F, is an electrochemiluminescence assay that was developed using serotypes 25 and 72 for preadsorption.
The IgG antibody response in early infancy is mainly of the IgG 1 subclass. After the PCV booster dose in the second year of life, IgG 2 antibodies also start to appear, and the response in adults to both pneumococcal polysaccharide and conjugate vaccines is mainly of the IgG 2 subclass. , In adults, some differences have been found in the IgG 1 :IgG 2 ratio elicited after vaccination with pneumococcal polysaccharide and various conjugate vaccines. However, the relevance of these findings for estimating the T-cell dependence of the response in infants and the functional activity of antibodies remains to be elucidated.
Avidity
Antibody avidity describes the strength with which an antibody binds to a complex antigen. Several methods have been described for determining relative antibody avidity to different types of antigens. In EIA techniques, the binding of antibody to the coating antigen may be prevented by competitive inhibition using decreasing concentrations of free antigen or, more usually, by eluting the antibody from the antigen by a dissociating agent, such as thiocyanate, urea, or diethylamine. Thiocyanate anion and urea interfere with the antibody-antigen binding primarily through disrupting the hydrophobic bonds; diethylamine is a protein-denaturing agent. Two avidity assays based on pneumococcal EIA and using sodium thiocyanate elution have been developed. Because the elution assays are based on dissociation of antibody-antigen complexes of low avidity, they allow the ranking of the antibodies by their avidity. The avidity of antibodies induced in infants after pneumococcal conjugate vaccination increases with time from the postprimary samples to prebooster samples. , , Experience with Hib conjugate vaccines suggest that measurement of antibody avidity is also useful in assessing the induction of memory.
Opsonophagocytic Activity
Because opsonin-dependent phagocytosis is the primary host defense mechanism against S. pneumoniae , a variety of techniques to measure the OPA of antibodies to pneumococci have been developed. , The intent of measuring OPA is to assess the function of antibodies rather than just their quantity because a proportion of antibodies measured in EIA have no functional activity. In support of the likely improved utility of OPA over EIA is the observation in animal models that OPA provides better correlation with protection than the concentration of IgG antibodies as measured by EIA. , In humans, sera from the Finnish otitis media trial, reanalyzed using an OPA assay, revealed that the OPA titers predicted the observed low efficacy for serotype 19A, whereas the enzyme-linked immunosorbent assay (ELISA) titers did not. Furthermore, the cross-reactive IgG response to serotype 19A in the seven-valent vaccine, which was not associated with protection against invasive pneumococcal disease, lacked significant OPA activity. Of importance, EIA results may have limited relevance in older adults, who often have more nonfunctional anticapsular antibodies than infants and children. Interest in OPA has increased in recent years as the primary assay to evaluate the functional activity of the immune response to PCV, and is now an essential part of the registration of new conjugate vaccines.
The classic OPA is an assay to determine the titers of sera that reduce the number of live bacteria by more than half through opsonophagocytosis. Early studies were performed using human polymorphonuclear leukocytes as the effector cells, but current assays have been adapted to use cultivable phagocytes (e.g., HL-60 or NB-4 cells). Romero-Steiner and associates developed an OPA assay for pneumococcal antibodies, in which they used rabbit complement and HL-60 cells as phagocytes. This method has become a reference against which other assays are compared. Additional OPA techniques, including radioisotopic, flow cytometric, microscopic, and viability (or killing) assays, have been tested. The killing OPA method, which measures the killing of live bacteria by effector cells, is regarded to be the most biologically relevant. , There is also more information about the validation and performance of the killing-type OPA than about the flow cytometric OPA. In general, the results obtained by the various techniques correlate well and an assessment of assays from six laboratories showed acceptable agreement between their assays.
With increasing need for high-throughput and automated functional assays, multiplexed methods for OPA have been developed. The Romero-Steiner assay has been validated to measure seven serotypes simultaneously, using fluorescent dye to quantify the bacterial killing. Another approach is to measure phagocytosis by flow cytometry, using fluorescently labeled bacteria or latex particles, and a multiplexed phagocytosis assay for three to four different serotypes has been developed using bacteria or latex beads coated with different polysaccharides. Still another type of multiplexed OPA is based on a killing-type OPA, using pneumococcal strains resistant to clinically irrelevant antibiotics. ,
The correlation between the IgG EIA antibody concentration and OPA results is relatively strong in immune sera , , , but is weaker at low antibody concentrations. , , , In general, the correlation between antibody concentration and OPA seems to be good in infants who have been immunized with conjugate vaccines, , but is weaker in unimmunized children and in adults. , , Correlations between EIA and OPA results are high for vaccine serotypes but lower for cross-reacting serotypes such as 6A and 19A. One reason for the poor correlation is that antibodies detected by standard EIA are not functionally active. , , As discussed below, a recent indirect cohort method was used to attempt to identify an aggregate correlate of protection using OPA. However, the variation in serotype-specific OPA values that correlated with protection was too great, and thus no such value could be defined.
Immunologic Memory
Memory is traditionally characterized by the presence of IgG antibody elicited in a previously naïve individual (such as an infant), by evidence of priming after immunization, or by affinity maturation. For the purposes of vaccine evaluation, the simplest method of demonstrating memory is often the increased concentration and IgG dominance of an antibody response after a booster dose. The increase of geometric mean concentration (GMC) in a primed individual with immunologic memory is much stronger than that seen if the same vaccines were given at the same age without precedent priming. ,
The maturation of antibody avidity and increased OPA have also been evaluated in studies of pneumococcal vaccines. , The avidity of antibodies is increased in children who receive a pneumococcal conjugate booster but not a polysaccharide vaccine booster. , This suggests that the conjugate triggers a T-cell response, but the T-cell-independent PPV only triggers the existing memory B cells. , ,
Antibody response may be higher after the polysaccharide booster than after the conjugate booster in adults. , Whether this has any clinical advantage is not clear. Theoretically, a conjugate booster may stimulate the generation and expansion of high-affinity clones of B memory cells, whereas use of a polysaccharide booster could result in depletion of the memory pool. Long-term (5-year) follow-up of children receiving polysaccharide vaccine did not, however, show reduction in PCV responses or the pool of polysaccharide-specific memory B cells. Evidence in support of this idea was found in a comparison of memory B cells following polysaccharide versus conjugate pneumococcal vaccination of adults. Therefore, some have argued that boosting with a conjugate may prove important for the persistence of immunity, although this has not been conclusively shown. It is also important to note that the use of the polysaccharide booster may be associated with more pronounced reactogenicity.
The response to the booster dose of either PCV or PPV is rather vigorous at 24 months of age, as seen in the increase in GMCs of anticapsular antibodies in samples taken 7–10 days (but not yet at 4 days) after immunization, provided the children had received primary doses with the conjugate. , , A comparison of a booster dose at 12 months of age between PCV13 and PCV10, given to those previously primed with the same vaccine in infancy, suggests that the memory B-cell response to PCV13 appeared superior to PCV10 for most serotypes common between the vaccines. The clinical relevance of this observation, if any, related to impact on carriage or duration of protection has not been established. The priming effect is less impressive in healthy adults, , maybe because adults often have already encountered the antigen and become primed, either through carriage of the bacterium or other, cross-reacting antigens. Some indications of priming can be seen also in patients with HIV, , after bone marrow transplant, , and sickle cell anemia.
Whether mucosal memory is induced by PCVs is still under examination. In one study from the United Kingdom, salivary IgA concentrations against pneumococcal serotypes were only slightly higher in the PCV7 vaccinated group than in the unvaccinated control group, but there was a significant increase in mucosal antibodies after a booster dose of polysaccharide. Mean-fold rises after the polysaccharide vaccine were lower for serotypes not included in the conjugate, suggesting that the conjugate vaccine had primed children for memory responses. However, two other studies found no clear differences in salivary antibody responses between children primed with conjugates and those who received their first pneumococcal vaccine during the second year of life. , Follow-up data from the Finnish Otitis Media Vaccine Efficacy Trial (FinOM), showed that PCV7 induced both salivary IgG and IgA responses to the vaccine serotypes. By 4–5 years of age, IgA concentrations were similar in both the vaccinated group and the nonvaccinated control group, most probably reflecting natural stimulation through repeated episodes of colonization.
Whether immunologic memory plays an important role in vaccine-induced protection against pneumococcal disease remains controversial. The experience with Hib immunization in infants supported the view that, in addition to the potent herd immune effect, priming may have provided some protection against disease because very young children were protected despite low antibody concentrations. However, priming and memory to the Hib polysaccharide induced by natural exposure in the prevaccine era or Hib conjugate vaccine , may not be sufficient for protection against disease, as the United Kingdom’s experience with Hib strongly suggests. To explain the observation of a rise in failures of the Hib vaccine between 1999 and 2003, the introduction of an acellular pertussis-containing Hib conjugate vaccine without a booster suggested that this vaccine primed well but was associated with lower immunogenicity. , In contrast, no such increase in failures was noted in countries that adopted this new vaccine but did not eliminate the booster dose. Similarly, studies of the serogroup C N. meningitidis conjugate vaccines in the United Kingdom demonstrate lower effectiveness 1 year after vaccination, suggesting that priming of the immune response and memory responses may not be sufficient to prevent invasive disease. It is possible that the lack of efficacy against serotype 1 pneumococcal disease in the PCV9 trials in Africa may also reflect failures in the second year of life after a primary vaccination schedule without a booster. Because of the observed delay in antibody response in primed subjects after exposure, a memory response may not be rapid enough to protect against an isolate that has the capacity to rapidly invade the bloodstream soon after establishing nasopharyngeal colonization. Thus, the role of immunologic memory in conferring long-term protection against invasive bacterial disease remains controversial.
Animal Models
In addition to in vitro assays, many animal models have been developed to study the immunogenicity and efficacy of pneumococcal conjugate and novel protein vaccines, including rodent (mice and rats), rabbit, chinchilla, and infant monkey models. In the immunogenicity studies, animals are bled before and after active immunization with an experimental vaccine to obtain sera or cells for various analyses. , , The functionality of antibodies can also be evaluated in animal models. , , , Animals may be passively or actively immunized and subsequently challenged with bacteria to determine protective efficacy. ,
Many different strains of pneumococci can be used for colonization and challenge studies, although there are marked serotype-dependent differences with respect to their capacity to colonize the nasopharynx or cause invasive disease in animal models. Various routes of challenge have been employed, in attempts to mimic the human situation of nasopharyngeal colonization, otitis media, bacteremia, pneumonia, and meningitis. In nasopharyngeal colonization studies, using clinical isolates for inoculation can result in sustained colonization for several days to weeks. Using the same challenge protocol but under anesthesia and with a larger inoculum (50 µL or more), mice aspirate the inoculum, which subsequently leads to lung infection and, depending on the strain, potential sepsis (particularly if lower numbered serotypes are used). Adult and infant mice and rats have also been used in passive-protection models to determine the protective capacity of human antipneumococcal antibodies against bacteremia, meningitis, or death. , , ,
In addition, special mention should be made of a human model of intentional pneumococcal exposure that permits an evaluation of the immune responses that are either associated with resistance to colonization or response to exposure. , Use of this model demonstrated the nasal inoculum required to establish colonization in healthy adults (50% colonizing dose estimated at 10 3 –10 4 colony-forming units [CFU]), defined the duration of colonization (range: 0–122 days) and analyzed the mucosal and systemic humoral responses to a single carriage event. Such studies could be particularly useful in the future for testing of vaccines that target carriage. Indeed, this model has been used for evaluation of the protective effect of PCV13 against carriage.
Infants and Young Children
PCVs were originally designed to be used as part of routine infant immunization programs; consequently, most clinical trials have focused on the safety and immunogenicity of candidate vaccines in infants. Various dosing schedules have been examined and the available literature on immunogenicity, efficacy, and effectiveness of different schedules has been systematically reviewed. Although aggressive schedules starting with vaccination at birth were studied, they have not been approved. In early infancy, passively acquired maternal antibodies interfere with the response to the first doses of conjugates, although there is no evidence of interference for any serotype after completion of the entire immunization series. In the Expanded Programme on Immunization (EPI) schedule, the conjugate is given at the ages of 6, 10, and 14 weeks. However, the schedule that has been studied most extensively is a three-dose primary immunization series either at 2, 3, and 4 months or at 2, 4, and 6 months of age, followed by a booster dose during the second year of life. Reduced vaccination schedules at 2, 4, and 12 months or at 3, 5, and 11–12 months of age have also been evaluated. Although three doses were originally considered necessary for an optimal immune response to conjugate vaccines, , recent studies indicate that even one dose of conjugate vaccine may be sufficient, at least in circumstances where carrier priming and early stimulation with high carriage of pneumococci is common. Direct studies comparing various schedules are few, but in general, some conclusions can be drawn: (a) a short interval between doses (i.e., 1 month) often results in lower antibody response than a longer one (i.e., 2 months) ; (b) for some serotypes (in particular serotypes 6B and 23F), reduced schedules may show significantly lower immunogenicity, when immunogenicity prior to the booster dose in a two-dose regimen is compared with a three-dose regimen , ; and (c) although a single dose in infancy is immunogenic, it is inferior to a two- or three-dose regimen for all serotypes. In this regard, further evaluation of the quality of antibodies and induction of memory, as well as of the determination of the optimal age of vaccination, would be of interest. , Although most of the early data on immunogenicity were derived from the seven-valent and nine-valent CRM 197 -conjugated vaccines (PCV7 and PCV9), data on other early PCVs (conjugated to diphtheria toxoid, tetanus toxoid, and OMPC) are available. Studies conducted for the licensure of H. influenzae protein D (PHiD-CV, PCV10), and the 13-valent CRM 197 conjugate vaccine suggest that they elicit antibody responses somewhat lower than PCV7 in infants and toddlers (see Table 47.3 and 47.4 on 13-year valent). , , , , , Thus, the addition of more serotypes contained in PCV10 or PCV13 results in a wider spectrum of immune response to the additional serotypes with generally minor compromise to the response against the seven serotypes in PCV7.
TABLE 47.3
Antibody Concentrations (GMCs, µg/mL) 1 Month After Last Dose of the Primary Series of Pneumococcal Conjugate Vaccines in Infants
| Company | Vaccine | Site/Population | N | Schedule | 1 | 3 | 4 | 5 | 6B | 7F | 9V | 14 | 18C | 19F | 23F | 6A | 19A | References |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Pfizer | PCV7 | U.S. | 88 | 2, 4, 6 months | – | – | 1.46 | – | 4.70 | – | 1.99 | 4.60 | 2.16 | 1.39 | 1.88 | – | – | |
| Pfizer | PCV7 | U.S. | 75 | 2, 4, 6 months | – | – | 1.34 | – | 2.14 | – | 1.23 | 5.04 | 1.88 | 1.52 | 1.20 | – | – | |
| Pfizer | PCV7 | Finland | 57 | 2, 4, 6 months | – | – | 1.70 | – | 2.00 | – | 2.50 | 6.30 | 3.60 | 3.30 | 2.50 | – | – | |
| Pfizer | PCV7 | Navajo and Apache, U.S. | 223 | 3 doses at age 6 weeks –7 months | – | – | 3.21 | – | 8.25 | – | 2.47 | 6.81 | 2.60 | 2.74 | 2.59 | – | – | – a |
| Pfizer | PCV7 | Israel | 715–782 | 2, 4, 6 months | 0.02 | 0.04 | 2.02 | 0.23 | 2.30 | 0.05 | 1.44 | 6.46 | 1.47 | 2.23 | 1.57 | 0.36 | 0.70 | |
| Pfizer | PCV9 | South Africa b | 63 | 6, 10, 14 weeks | 7.55 | – | 4.09 | 5.79 | 1.76 | – | 3.35 | 3.62 | 4.55 | 6.02 | 3.15 | – | – | |
| Pfizer | PCV9 | South Africa c | 30 | 6, 10, 14 weeks | 3.45 | – | 2.77 | 3.32 | 1.21 | – | 2.36 | 2.23 | 1.87 | 3.59 | 1.78 | – | – | |
| Pfizer | PCV9 | The Gambia | 217 | 2, 3, 4 months | 6.94 | – | 4.90 | 5.84 | 4.93 | – | 4.07 | 4.45 | 4.89 | 2.91 | 2.85 | – | – | |
| Merck | PncOMPC | Finland | 376 | 2, 4, 6 months | – | – | 3.45 | 0.35 | – | 1.79 | 3.23 | 1.02 | 3.19 | 0.67 | – | – | ||
| GSK | PncPD | Czech Republic | 140 | 3, 4, 5 months | 1.58 | 3.78 | 2.16 | 1.92 | 0.62 | 2.34 | 1.60 | 3.00 | 1.49 | 2.60 | 0.90 | – | – | |
| GSK | PCV10 | Finland, France, Poland | 1107 | 2, 3, 4 months | 1.05 | – | 1.45 | 1.70 | 0.33 | 1.72 | 1.32 | 2.90 | 1.66 | 1.84 | 0.53 | – | – | |
| GSK | PCV10 | Germany, Poland, Spain | 173 | 2, 4, 6 months | 1.00 | – | 1.70 | 1.69 | 0.71 | 2.25 | 1.58 | 3.76 | 2.34 | 3.81 | 0.96 | – | – | |
| GSK | PCV10 | Philippines | 285 | 6, 10, 14 weeks | 3.23 | – | 4.96 | 4.87 | 1.19 | 4.84 | 4.04 | 6.45 | 11.56 | 10.46 | 2.23 | – | – | |
| Pfizer | PCV13 | U.S. | 252 | 2, 4, 6 months | 2.03 | 0.49 | 1.31 | 1.73 | 2.10 | 2.57 | 0.98 | 4.74 | 1.37 | 1.85 | 1.33 | 2.19 | 2.07 | |
| Pfizer | PCV13 | U.S. | 258–264 | 3, 5 months | 2.30 | 1.15 | 2.38 | 1.27 | 0.41 | 2.06 | 1.68 | 2.84 | 1.72 | 3.42 | 0.61 | 1.17 | 2.87 | |
| Pfizer | PCV13 | Germany | 285 | 2, 3, 4 months | 1.83 | 1.55 | 2.18 | 1.31 | 0.98 | 2.59 | 1.65 | 4.14 | 1.94 | 1.73 | 1.26 | 1.33 | 3.26 | |
| Pfizer | PCV13 | U.K. | 107 | 2, 4 months | 1.69 | 0.63 | 1.37 | 0.95 | 0.26 | 2.14 | 0.87 | 1.83 | 1.37 | 2.38 | 0.53 | 0.86 | 1.90 | |
| Pfizer | PCV13 | Israel | 741–765 | 2, 4, 6 months | 2.08 | 0.97 | 2.16 | 1.38 | 2.26 | 3.34 | 1.40 | 5.72 | 1.49 | 2.90 | 1.13 | 2.53 | 1.81 | |
| Serum Institute of India | PCV10 | Gambia | 1458 | 6,10,14, weeks | 4.29 | 1.65 | 1.21 | 2.97 | 1.31 | 5.20 | 4.35 | 1.58 | 1.00 | 1.64 |
–, Indicates that the serotype was not present in the vaccine; GMCs, geometric mean concentrations; GSK, GlaxoSmithKline; OMPC, outer membrane protein conjugate; PncPD, pneumococcal protein D (conjugate); Ref, reference.
Data derived from pivotal trial (either efficacy trials or trials used for licensure of new vaccines after PCV7 introduction). Note that the studies were conducted in different populations, during different time periods, and the antibody assays were performed in different laboratories; consequently, results from separate studies are not directly comparable. 714
TABLE 47.4
U.S. Advisory Committee on Immunization Practices Recommended Schedule for Administering Doses of 13-Valent Pneumococcal Conjugate Vaccine (PCV13) to Children <24 Months by PCV Vaccination History and Age
| Age at Current Visit (months) | Vaccination History: Total Number of PCV13 Doses Received Previously | Recommended PCV13 Regimen a |
|---|---|---|
| 2–6 | 0 | 3 Doses, 8 weeks apart; fourth dose at age 12–15 months |
| 1 | 2 Doses, 8 weeks apart; fourth dose at age 12–15 months | |
| 2 | 1 Dose, 8 weeks after the most recent dose; fourth dose at age 12–15 months | |
| 7–11 | 0 | 2 Doses, 8 weeks apart; third dose at 12–15 months |
| 1 or 2 before age 7 months | 1 Dose at age 7–11 months, with a second dose at 12–15 months, ≥8 weeks later | |
| 12–23 | 0 | 2 doses, ≥8 weeks apart |
| 1 before age 12 months | 2 doses, ≥8 weeks apart | |
| 1 at ≥12 months of age | 1 dose, ≥8 weeks after the most recent dose | |
| 2 or 3 doses before age 12 months | 1 dose, ≥8 weeks after the most recent dose | |
| 24–59 | Any incomplete schedule in a healthy child | 1 dose |
a Minimum interval between doses is 8 weeks except for children vaccinated at <1 year of age, for whom minimum interval between doses is 4 weeks. 716
In general, anticapsular antibody GMCs increase 5-fold to 10-fold after the initial series, relative to the preimmunization concentrations. The antibody concentrations achieved are usually only sustained for a few months and decline during the second year of life to about the preimmunization levels. However, a dose of the pneumococcal vaccine, either polysaccharide or conjugate, administered during the second year of life to children primed with any of the conjugates, generally induces an approximately 10-fold increase in antibody concentrations. Typical ways to describe the immunogenicity results are reverse cumulative distribution curves ( Fig. 47.2 ) or radial diagrams (“logograms”) ( Fig. 47.3 ). Table 47.5 compiles representative immunogenicity data after the primary immunizations from selected pivotal trials (either efficacy trials or trials used for licensure of new vaccines after PCV7 introduction [relevant to PCV10 and PCV13]).
TABLE 47.5
Composition of the Five Currently Licensed Pneumococcal Conjugated Vaccines and the Original PCV7 a
| Vaccine | Commercial Name | Polysaccharide Content (µg) and Protein Carrier | Adjuvant | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 3 | 4 | 5 | 6A | 6B | 7F | 9V | 14 | 18C | 19A | 19F | 23F | 22F | 33F | 8 | 10A | 11A | 12F | 15B | |||
| PCV7 (CRM 197 conjugate) | Prevnar/Prevenar | (–) | (–) | 2.0 | (–) | (–) | 4.0 | (–) | 2.0 | 2.0 | 2.0 | (–) | 2.0 | 2.0 | Aluminum phosphate | |||||||
| PCV13 (CRM 197 conjugate) | Prevnar13/Prevenar13 | 2.2 | 2.2 | 2.2 | 2.2 | 2.2 | 4.4 | 2.2 | 2.2 | 2.2 | 2.2 | 2.2 | 2.2 | 2.2 | Aluminum phosphate | |||||||
| PCV10 (serotypes 1, 4, 5, 6B, 9V, 14, 23F conjugated to nontypeable Haemophilus influenzae -derived protein D; serotype 18C conjugated to tetanus toxoid; serotype 19F conjugated to diphtheria toxoid) | Synflorix | 1.0 | (–) | 3.0 | 1.0 | (–) | 1.0 | 1.0 | 1.0 | 1.0 | 3.0 | (–) | 3.0 | 1.0 | Aluminum phosphate | |||||||
| PCV10 (CRM197 conjugate) | Pneumosil | 2 | 2 | 2 | 4 | 2 | 2 | 2 | 2 | 2 | 2 | Aluminum phosphate | ||||||||||
| PCV15 (CRM197 conjugate) | Vaxneuvance | 2 | 2 | 2 | 2 | 2 | 4 | 2 | 2 | 2 | 2 | 2 | 2 | 2 | 2 | 2 | ||||||
| PCV20 (CRM197 conjugate) | Prevnar20 | 2.2 | 2.2 | 2.2 | 2.2 | 2.2 | 4.4 | 2.2 | 2.2 | 2.2 | 2.2 | 2.2 | 2.2 | 2.2 | 2.2 | 2.2 | 2.2 | 2.2 | 2.2 | 2.2 | 2.2 | |
a Information as described in product monographs presented for licensure.
Two doses of PCV7 given at either 2 and 4, 3 and 5, or 4 and 6 months of age as a primary series, together with a booster at 11–12 months, seem to induce an immune response and induction of immunologic memory comparable to those noted with the standard four-dose schedule in both full term and preterm infants. , , Thus, the marked booster response, after two doses in infancy, even for the least immunogenic serotypes, has led many countries to adopt a two-dose infant series and a booster rather than the initially licensed schedule of three infant doses and a booster.
Once herd protection has been induced in a widely PCV vaccinated infant population, consideration has moved to the minimum number of doses required to maintain herd protection. This is a cost-effectiveness consideration and also one in which the number of immunizations given to infants at each point in the schedule may be reduced. In the initial study 213 infants were randomized 1:1 to receive either PCV13 at 2 and 4 months or a single dose at 3 months, with all infants receiving a 12-month booster dose. While the immunogenicity was lower for all serotypes after the primary series, except for serotype 3, for which neither regiment induced a strong response, there was enhanced immunogenicity in the one dose primary group to serotypes 1, 4, 14, and 19F after the booster with a reduced response to serotypes 6A, 6B, 18C, and 23F and no significant difference in the remaining 5 serotypes. Following this trial, the UK adopted the 1+1 schedule in January 2020. The first developing country study of the 1+1 schedule expanded the UK study to 1695 infants in South Africa and included both the Pfizer PCV13 and GSK PCV10 vaccines. Investigators randomized the single dose to either 6 weeks or 14 weeks of age, while the control group received the first dose at both 6 and 14 weeks of age. All trial participants received their booster dose at 9 months of age. For PCV13, the data were remarkably similar to the UK data with only serotypes 6B and 23F falling below a 50% noninferiority margin after the booster dose for the 1+1 group and significantly increased titers post boost for serotypes 1 and 19F. In general, the preboost titers were higher when the single dose primer was given at 14 rather than 6 weeks of age. The GSK data were similar, with no serotypes below a 50% noninferiority margin post boost. Further data will be forthcoming from a similar trial in India and a cluster randomized trial from Na Trang, Vietnam of both 1+1 and a regimen of a single booster dose alone. The WHO has recommended the two primary doses plus a booster series as an alternative to the three-dose primary series typically used in EPI schedules, but has yet to weigh in on the 1+1 schedule.
In addition to the vaccination schedule and number of doses, the characteristics of the target population also affect the immune response. Direct comparisons are difficult, however, because both the vaccination schedules and the laboratories performing the serologic assays are usually not consistent in all the studies. Similar comparisons were conducted with PCV7 and PCV9 vaccines containing either seven or nine serotypes, which were evaluated in studies conducted in the United States, Finland, and South Africa and showed greater overall immunogenicity in South Africa than in Finland and the United States. ,
Immune responses to PCVs are clearly associated with vaccine protection, but the magnitude varies with geographic region and serotype. , Children enrolled in studies from Asia, Africa, and Latin America had significantly higher antibody responses compared to those from Europe and North America given similar vaccination schedules. The reasons for these differences could derive from many factors, such as the differences in the nasopharyngeal microbiome of the children, priming effect of tetanus toxoid given to pregnant women, earlier pneumococcal acquisition, differences in concomitant vaccine administration, and genetic differences among populations. Similar differences were noted in immune responses to Hib conjugate vaccines in earlier studies.
The IgG responses of children with invasive pneumococcal disease (IPD) have been measured and values compared between those previously immunized with PCV versus unimmunized children. Vaccinated children with IPD are less likely to have greater than 0.35 µg/mL of antibody to the infecting serotype compared with other serotypes. These children with IPD proved hyporesponsive to their infecting serotype in subsequent exposure to PCV, although recurrent disease was rare (0.2%).
Preterm Infants
Preterm infants respond to the PCV satisfactorily, although the immunogenicity of the vaccine in preterm infants is usually lower than in term infants. , , The Advisory Committee on Immunization Practices (ACIP) in the United States recommends that PCVs be given to preterm infants at their chronologic age, consistent with the schedule recommended for full-term infants. It may be particularly important to offer a timely booster dose of PCV13 to such infants. , The response of preterm infants after booster depends on the number of doses in the primary series; infants who received only two doses in the primary series were less likely to have antibody levels above a threshold correlated with protection after the primary series than children who received a three-dose primary series but were more likely to have levels above the threshold after boost.
Neonatal Immunization
Neonatal administration of PCV does not lead to tolerance of subsequent doses and postprimary responses are similar or slightly lower than PCV administered to older infants. The somewhat lower PCV immunogenicity at birth, and relative scarcity of neonatal pneumococcal infections, as well as induction of herd protection to protect infants younger than 6 weeks of age has not translated into use of PCV at birth.
Impact of Maternal Immunization on Immunogenicity in Infants
In a randomized trial of PCV9 in pregnant women, administered at 30–35 weeks of pregnancy, serotype-specific IgG was significantly raised in infant cord blood compared to infants whose mothers received placebo. Infants then received PCV7 at 2, 4, 6, and 12 months. At 7 months of age, the IgG responses to the initial three-dose series of PCV7, were significantly reduced for all serotypes among infants of immunized mothers, compared to infants of mothers who received placebo. Robust booster responses to PCV7 were measured 1 month after the 12-month dose. The mean levels of IgG remained lower in infants of vaccinated mothers compared to those born to unimmunized control mothers for serotypes 6B, 18C, 19F, and 23F. These data suggest that high levels of passively acquired maternal antibody suppress the primary responses to those serotypes in their infants. Responses to serotypes 1 and 5, not given to the infants, remained elevated compared to infants of unimmunized control mothers through 6 and 13 months of age. In addition data from infants born to mothers receiving tetanus, diphtheria, and pertussis (Tdap) during pregnancy had reduced responses to CRM based PCV conjugate compared to historical controls whose mothers did not receive Tdap.
Older Children, Adults, and the Elderly
GMCs for most serotypes in infants 1 year of age receiving two doses of PCV and older children receiving a single dose of PCV7 were comparable to the immune responses of infants who received three doses of conjugate in the Northern California Kaiser Permanente (NCKP) efficacy study. Based on these data, the official recommendation in the United States is to use PCV13 at 2, 4, 6, and 12–15 months of age. For children 12–23 months of age, two doses are recommended at least 2 months apart, and for those 24 months through 9 years of age, one dose is considered sufficient. PCV13 vaccination is recommended for all children 2–59 months of age and those 60–71 months of age with underlying medical conditions (see Table 47.5 ). Children 2–18 years of age with certain underlying medical conditions are recommended to receive PPV23 following their full schedule of PCV13. In February 2013, the ACIP recommended routine use of PCV13 for children 6–18 years of age with immune-compromising conditions, functional or anatomic asplenia, CSF leaks, or cochlear implants who have not recently received PCV13, regardless of whether they previously received PCV7 or PPV23. The recommendation was for a single PCV13 dose first, followed 8 or more weeks later by a PPV23 dose. A second PPV23 dose is recommended 5 years after the first PPV23 for children with anatomic and functional asplenia, HIV infection, or other immunocompromising condition. If these children previously received one or more doses of PPV23, they should be given a single PCV13 dose 8 or more weeks after last PPV23 dose. These recommendations are based on immunogenicity studies in both immunocompetent and immunocompromised children 6–18 years of age.
To date, there is some evidence that conjugate vaccines are more immunogenic than the PPV vaccine in adults, although this has not been consistently shown. In early studies, a five-valent PCV vaccine elicited a greater response than PPV to some serotypes (6B, 18C, and 23F) but not others (14 and 19F). , In more recent studies, the IgG and opsonophagocytic responses to PCV7 and PCV13 were either similar or higher than those to PPV. In a dose ranging study by Jackson and colleagues, administration of a 1-mL PCV7 dose (double the infant dose, so 4 µg of all serotypes except 8 µg of serotype 6B) demonstrated an enhanced immune response, but no further immunologic advantage was seen with a 2-mL PCV7 dose (fourfold greater than the infant dose). In adult Alaskan natives, PCV did not induce higher IgG and opsonophagocytic responses when compared with PPV and provided no benefit when given before PPV. There is also some evidence that the baseline level of antidiphtheria antibodies is associated with the response to the PCV7 vaccine: adults with higher diphtheria antibody levels have higher responses to the conjugate. Repeated PCV doses in adults seem to induce comparable immunogenicity as the first dose. , Receipt of PCV 6 months or 1 year after PPV led to a diminished response to PCV compared with PPV-naïve adults. , The order of immunization of adults with the conjugate versus polysaccharide vaccine appears to matter, with conjugate vaccination first conferring a booster response to a later dose of polysaccharide, while an initial dose of polysaccharide vaccine leads to a reduced response to PCV. ,
Because most adults have preexisting antibodies to pneumococcal polysaccharides and to the carrier protein, the initial vaccination with the conjugate acts as a booster.
Maturation of pneumococcal antibody avidity is prominent after the vaccination of infants, but the presence of mature antibodies before vaccination of adults results in only modest increases in avidity. In a study of different pneumococcal conjugates, the avidity index was stable in adults, whereas clear increases occurred in children after both primary and booster immunizations with PCV. Other reports have noted no avidity maturation in elderly subjects vaccinated with PPV, or young adults vaccinated with either PPV or PncD/T conjugate.
Although both IgG and OPA responses to the PncT conjugate and to the polysaccharide vaccine appeared to be similar in a group of patients with chronic obstructive lung disease and in healthy adults ; in patients recovering from pneumonia, there was better persistence of antibody after PCV7 than after PPV23 when assessed 6 months after immunization. This study also found little utility in combining PCV and PPV regimens. In patients with chronic obstructive pulmonary disease there was an improved persistence at 1–2 years postimmunization of functional activity for the common types except 19F in individuals given PCV7 compared to those given PPV23. Improved OPA persistence was also seen in hospitalized frail elderly a year post PCV followed by PPV23 given 6 months later, compared to 1 year post-PPV23 alone, but these data are confounded by the shorter time post-PPV23 in the combined immunization group. These data do suggest that in nonimmunocompromised adults, there may be an advantage of PCV over PPV, particularly in the duration of antibody response and the response to revaccination. A conjugate vaccine of comparable valency to PPV23 would thus likely be a significant advance, but for the reasons outlined earlier would pose technical and financial challenges.
Immunocompromised Individuals
In the absence of immunogenicity data in immunocompromised individuals, the immunogenicity cutoffs for immunocompetent individuals have been applied to the immunocompromised population.
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here