Reconsidering Guidelines on the Use of Pneumococcal Vaccines in Adults 65 Years or Older

The recent adult pneumococcal vaccine recommendations issued by the Advisory Committee on Immunization Practices (ACIP) and endorsed by the Centers for Disease Control and Prevention are well intentioned but misguided, complex, and not ready for widespread adoption.1 In August 2014, the ACIP1 recommended the newer pneumococcal conjugate vaccine 13 (PCV13) ($152 per dose2) for adults 65 years or older followed 6 to 12 months later by the older pneumococcal polysaccharide vaccine 23 (PPSV23) ($72 per dose2) (Box). The PCV13 covers 12 of 23 pneumococcal serotypes in PPSV23 plus an additional serotype, and contains conjugate protein, which renders a more robust immune response in individuals with immature or impaired immune systems. In the United States, the conjugate vaccine has been recommended for several years for young children and for patients who are immunocompromised. In its 2014 report describing the new guidelines, the ACIP1 prominently cited the Community-Acquired Pneumonia Immunization Trial in Adults (CAPiTA), a randomized trial of PCV13 among 85 000 adults 65 years or older, but it was not until March 2015 that the results of CAPiTA were published.4 To place the CAPiTA trial in context, it is important to recognize that the older PPSV23 vaccine has established efficacy against pneumococcal disease. A recent Cochrane Collaboration analysis5 of 11 randomized trials involving 36 489 participants 16 years or older concluded that PPSV23 was 74% (95% CI, 55%-86%) effective against invasive pneumococcal disease and 74% (95% CI, 54%-85%) effective against confirmed pneumococcal pneumonia. However, PPSV23 is far from perfect. There have yet to be any high-quality randomized trials focusing on PPSV23 in elderly individuals, and the available data on the elderly population are at times conflicting (although overall, lower-quality observational data5 suggest 68% [95% CI, 53%-78%] effectiveness against invasive pneumococcal disease in elderly individuals). Studies1 have also raised questions about the durability of the immune response to PPSV23. In addition, since Pneumococcus is responsible for only a fraction of the cases of pneumonia, its effect on overall pneumonia rates, including nonpneumococcal pneumonia, is modest (estimated efficacy, 28%; 95% CI, 7%-44%).5 Despite these limitations, vaccination with PPSV23 is the standard of care for individuals 65 years or older in many countries, including the United States. Thus, the key issue is whether PCV13 adds value when given instead of, or in addition to, PPSV23. Unfortunately, CAPiTA did not answer the key question. Pfizer, the manufacturer of PCV13, conducted the study in the Netherlands,4 where adults do not routinely receive PPSV23, and compared PCV13 with placebo. The PCV13 vaccine performed well relative to placebo, although not overwhelmingly so. During the 4-year study, the efficacy of PCV13 against invasive pneumococcal disease was a modest 52% (95% CI, 22%-71%), and against confirmed pneumococcal pneumonia, its efficacy was just 31% (95% CI, 10%-47%). Moreover, vaccination with PCV13 did not significantly reduce the overall rate of community-acquired pneumonia, nor did it reduce mortality due to pneumococcal disease. Because CAPiTA included only adults 65 years and older and the Cochrane Review5 included individuals as young as 16 years, it would be misleading to directly compare efficacy rates for PCV13 in CAPiTA with the efficacy rates for PPSV23. Another notable limitation of CAPiTA is that it took place in the Netherlands before childhood pneumococcal vaccination was well established. Because children are important reservoirs for Pneumococcus, childhood vaccination reduces pneumococcal disease in all age groups (ie, herd immunity). Since routine childhood pneumococcal vaccination began in the United States in 2000, there has been a 90% reduction in vaccine-type invasive pneumococcal disease and a 50% reduction in overall disease among all age groups.6 Childhood vaccination in the Netherlands did not begin until 2006, which was 2 years before CAPiTA started and not in sufficient time for maximal herd protection. Thus, the results likely overstate the benefits of PCV13 relative to settings with longstanding childhood PCV13 vaccination. Still, the ACIP moved forward with its recommendation to include both PCV13 and PPSV23 in the vaccination routine, promising to reevaluate this recommendation in 2018. The committee chose this reevaluation date because, although US children began receiving a pneumococcal conjugate vaccine covering 7 serotypes (PCV7) in 2000, the newer PCV13, which covers 6 additional strains, was not routinely used until 2010. By 2018, herd immunity for these additional 6 strains should be well established. In the meantime, we are concerned that the guidelines from the ACIP and endorsed by the CDC will be comVIEWPOINT