Prevention of Pneumococcal Infection With Vaccines

THE FIRST VACCINES TO PREVENT PNEUMOCOCCAL INfections, crude preparations of killed bacteria, were developed by Sir Almroth Wright in 1911 to try to alleviate the high mortality and morbidity among gold miners in South Africa. Discovery that antibodies against purified polysaccharides of the capsular surface of pneumococci were protective led to development of polysaccharide vaccines that were marketed in the 1940s. These vaccines were commercial failures because the advent of antimicrobials led to a perception that pneumococcal infections were no longer a major threat. Subsequent evidence of the persistence of significant morbidity from pneumococcal infections, as well as mortality rates of 25% to 30% in patients with invasive (including bacteremic) pneumococcal infections despite early treatment with antimicrobials, led to redevelopment of a polysaccharide vaccine, approved in the United States in 1977, that contained 14 of the more than 90 serotypes of pneumococci (responsible for about 80% of invasive pneumococcal infections in the United States). In 1983 the current expanded formulation of the vaccine that contains polysaccharides of 23 pneumococcal serotypes (PPSV23) was introduced. Current recommendations are for immunization of persons 2 years and older with conditions that put them at increased risk of serious pneumococcal infections and all persons 65 years and older with PPSV23. Although there is good evidence that PPSV23 has at least moderate effectiveness in preventing invasive pneumococcal infections, its effectiveness in preventing nonbacteremic pneumococcal pneumonia (NPP) appears to be poor. Compared with bacteremic pneumococcal pneumonia, among adults NPP is thought to be at least 10 times more common and is responsible for considerably more morbidity and mortality. Children younger than 2 years used to have among the highest incidence of invasive pneumococcal infections. Polysaccharides, of which PPSV23 is composed, are T-cell–independent immunogens that have poor immunogenicity in young children. However, by bonding (conjugating) the capsular polysaccharides to certain proteins, the conjugated polysaccharide antigens are processed as T-cell–dependent immunogens to which even infants have an effective immunologic response. In addition, protein-polysaccharide conjugate vaccines induce immunologic memory with associated anamnestic (booster) antibody responses upon reexposure to the antigens and likely provide longer-lasting immunity than polysaccharide vaccines. The success of conjugate vaccines against Haemophilus influenzae type b, formerly the most common cause of bacterial meningitis in the United States but virtually eliminated since introduction of this vaccine for infants in 1988, encouraged development of conjugate vaccines against pneumococci. In 2000, a 7-valent conjugate vaccine against pneumococcus (PCV7) was introduced for routine childhood immunization. Although this vaccine was effective in reducing invasive disease against serotypes included in the vaccine, there was a much smaller but still significant increase in the incidence of disease caused by some serotypes not contained in PCV7 (notably serotype 19A). In 2010, PCV7 was replaced by a conjugate vaccine that includes 13 serotypes of pneumococci (PCV13) and is expected to further reduce the incidence of pneumococcal infections. The Food and Drug Administration recently approved PCV13 for use in adults. Work continues on development of vaccines based on other antigens that are shared by, and presumably could provide protection against, all serotypes of pneumococci. In this issue of JAMA, Smith and colleagues report that substitution of PCV13 for PPSV23 for routine immunization of adults would be more cost-effective per qualityadjusted life-year (QALY) gained (ie, $28 900/QALY gained compared with $34 000/QALY gained), consistent with similar analyses conducted in Europe. In addition, a single