Determining the Optimal Pneumococcal Vaccination Strategy for Adults: Is There a Role for the Pneumococcal Conjugate Vaccine? FREE TO VIEW

Despite a decline in the incidence of invasive pneumococcal disease in adults after the introduction of the 7-valent pneumococcal protein conjugate vaccine (PCV7) for use in children, there are approximately 24,000 cases per year of invasive pneumococcal disease among the elderly in the United States.¹ Nonbacteremic pneumococcal pneumonia remains a much more common problem, accounting for approximately 30% of all patients hospitalized with community-acquired pneumonia in which a cause is defined,² amounting to several hundred thousand of the approximately 1.3 million cases per year in the United States.³ The 23-valent pneumococcal polysaccharide vaccine (PPV23) has been available since 1983 and provides partial protection against invasive pneumococcal disease in immunocompetent elderly individuals,⁴ but has not been definitively shown to impact the overall risk of pneumonia in the elderly^5-8 or younger adults with comorbidities.^8,9

On February 24, 2010, the US Food and Drug Administration approved a 13-valent pneumococcal protein conjugate vaccine (PCV13) for use in children. The availability of PCV13 creates a unique dilemma for clinicians who care for adult patients at increased risk of pneumococcal disease, as it is unusual to have availability of a new pediatric vaccine for a disease that is important in adults. Further complicating the situation is the lack of clear evidence regarding the relative benefits in adults of PPV23 compared with PCV13. However, several lines of evidence suggest the possibility that PCV13 may be more efficacious in adults than PPV23. An additional concern is that prior receipt of the PPV23 may result in a blunted response to a pneumococcal conjugate vaccine. This commentary briefly reviews the data regarding the efficacy of PPV23 and pneumococcal protein conjugate vaccines in adults and discusses the policy implications of these data. For those who wish more detailed information about any of the various pneumococcal vaccines, several excellent reviews are available.^4,8,10-12

Although randomized controlled trials would provide the most definitive data about the relative efficacy of the PCV13 and PPV23 vaccines, performing such trials is extremely difficult and expensive. Sputum cultures and the urinary antigen tests for Streptococcus pneumoniae have less than optimum sensitivity; thus, documenting an organism-specific effect is quite challenging. In addition, pneumococcal pneumonia is a relatively uncommon outcome, except when considering large populations, so randomized trials of pneumococcal vaccine must be extremely large to have adequate statistical power. Although cohort and case-control studies are cheaper and logistically easier, they introduce bias from known and unknown confounders.

Because of these issues, the development of surrogate markers for vaccine efficacy has been critical. Enzyme-linked immunosorbent assay (ELISA) can be used to measure antipneumococcal IgG antibodies,¹³ giving reproducible results. However, there is no consensus regarding the level of the antibody that provides protection against infection in adults nor what defines an appropriate vaccine response. Although older adults develop antibody titers that are similar to those in their younger counterparts, these antibodies have reduced function¹⁴ and ELISA cannot distinguish between functional and nonfunctional antibodies.¹⁵ Antibodies against pneumococcal capsular polysaccharide function by opsonizing bacteria, which allows for phagocytosis by host immune cells. Opsonophagocytic killing (OPK) activity measures the level of serum antibody function in vitro with a phagocytic cell line and has been shown to correlate with immune protection in animal studies.¹⁶ Measures of OPK have also been shown to correlate with protection better than ELISAs for otitis media in children.¹⁷ However, there have been no studies correlating OPK assay results with protection in adults. Unlike ELISA, OPK does not yet have a reference standard.

Approximately 67% of individuals ≥ 65 years old in the United States have received at least one dose of PPV23.¹⁸ Initial studies of the vaccine in South African gold miners¹⁹ and Papua New Guinea highlanders²⁰ demonstrated that it is effective in preventing pneumococcal pneumonia and invasive pneumococcal disease in healthy young adults. Subsequent randomized trials have generally failed to demonstrate that PPV23 reduces all-cause⁴ or pneumococcal pneumonia^5,9,21 in older adults, suggesting suboptimal efficacy in preventing pneumococcal pneumonia. An exception is a recently published placebo-controlled randomized trial of PPV23 in Japanese nursing home patients, which demonstrated apparent protection against pneumococcal pneumonia and pneumonia-related mortality, but not all-cause mortality.²² An accompanying editorial raised concerns about the lack of a standardized methodology for interpreting the chest radiographs, the overall high rate of pneumonia diagnosed in these patients, and the importance of a decreased pneumonia-related mortality without a reduction in all-cause mortality.²³ Observational studies in elderly patients have given disparate results; some have suggested effectiveness in preventing pneumonia, but most have not.^4,6 Although some observational studies have demonstrated a protective effect with respect to hospital admissions and outcomes,²⁴ there has been concern that the healthy user effect might explain the improved outcomes associated with vaccination in some observational studies.⁹ In contrast to the disappointing experience with pneumonia, observational studies have consistently shown effectiveness in preventing invasive pneumococcal disease in healthy elderly individuals, with an overall vaccine effectiveness of approximately 60%.^6,25,26 There is no evidence that PPV23 protects against pneumococcal infections in adults who are immunosuppressed.^4,6,8,25

Because plain polysaccharide vaccines such as PPV23 are poorly immunogenic in infants, protein conjugate vaccines were developed, which link the polysaccharide antigen to a nontoxic protein carrier, thereby increasing the immune response to the polysaccharide. In young children, PCV7 is highly effective against pneumococcal bacteremia due to vaccine strains, and introduction of PCV7 into the childhood vaccination schedule led to a markedly decreased incidence rate of pediatric invasive pneumococcal disease.^27,28 In addition, in children, PCV7 appears to result in significant protection against nonbacteremic pneumococcal pneumonia and acute otitis media due to S pneumoniae, something not seen with the PPV23.¹¹ Finally, because of decreased nasopharyngeal carriage of pneumococcus in children after the release of PCV7, rates of invasive pneumococcal disease in adults also declined after PCV7 was released,²⁸ consistent with a herd-protective effect.

Several studies have suggested that PCV7 may also elicit a superior immune response in adults when compared with PPV23. Jackson et al²⁹ reported higher postvaccination antibody levels in previously vaccinated healthy patients ≥ 70 years of age using a 1.0-mL PCV7 dose, whereas de Roux³⁰ demonstrated similar results in a vaccine-naive elderly population using the traditional pediatric dose (0.5 mL). More recently, PCV7 (1.0 mL) was shown to induce a superior response in patients with COPD when compared with PPV23.³¹ Although both the PCV7 and PPV23 groups showed statistically significant increases in postimmunization antibody concentrations, PCV7 vaccination resulted in responses that were significantly higher for all but serotype 14 and 19A. OPK was generally higher after PCV7.^29,30 Although a recent study by Goldblatt et al³² found that PCV7 given to elderly patients resulted in higher antibody levels than PPV23 for only three of seven serotypes, they did not measure OPK, so the level of functioning antibody could not be compared, and they only used a single pediatric dose of the vaccine. There are limited comparative data in adults with respect to clinical outcomes after vaccination with PPV23 or PCV7. However, in Uganda, a randomized controlled trial of a conjugate vaccine (PCV7 0.5 mL given twice, with a 1-month interval) given to HIV-infected older children and adults following hospitalization for invasive pneumococcal disease demonstrated a 74% reduction in recurrent invasive pneumococcal disease,³³ whereas a prior randomized controlled trial with PPV23 in the same area was not protective.³⁴ Overall, the available data suggest that PCV7, particularly when administered at twice the usual pediatric dose, elicits a more robust functional immune response in adults than does PPV23.

A diminished antibody response to a subsequent polysaccharide-containing vaccine after initial vaccination with an unconjugated polysaccharide vaccine has been observed with a variety of vaccines, including meningococcal serogroup C vaccine,³⁵ and in recent comparative trials of sequential vaccination with PPV23 and PCV7.^30,36 The mechanism of this phenomenon, termed immune hyporesponsiveness, is unclear. One author has suggested that it may be because polysaccharide vaccines induce memory suppressor T cells,³⁷ and another suggested that it may be due to depletion of memory B cells after polysaccharide vaccination.³⁰ In one study, subjects who received PPV23 followed by PCV7 (0.5 mL) 1 year later had threefold lower ELISA and OPK than those who received PCV7/PCV7 or PCV7/PPV23 vaccination regimens.³⁰ In a study of PCV7 (1 mL) vaccination in patients with COPD, although all subjects who had been previously vaccinated had received PPV23 > 5 years prior to enrollment, immunologic hyporesponsiveness was still apparent.³¹ In two recent studies, patients who had received PPV23 3 to 5 years previously had modestly lower total antibody (30 and 60 days) and OPK responses (30 days) after revaccination with PPV23 than did vaccine-naive recipients, although these differences were not statistically significant for most serotypes.^37,38 These differences were no longer apparent at 5 years. In contrast, prior PCV7 (0.5 mL) vaccination does not induce hyporesponsiveness to subsequent immunization with either PCV7 or PPV23.^30,32 Because PCVs do not appear to inhibit subsequent responses to either conjugated or unconjugated vaccines, repeated vaccinations over time may greatly extend the window of protection of adults, in contrast to the current strategy of a one-time administration of PPV23 to targeted high-risk adults and those ≥ 65 years of age, a strategy that may blunt the more robust response to a subsequent conjugate vaccine.

A placebo-controlled trial of PCV13 (0.5 mL) in subjects ≥ 65 years of age is currently being performed in the Netherlands, with anticipated enrollment of 85,000 subjects.³⁹ Although it will not directly compare PCV13 to PPV23, the study is likely to provide critical information regarding the protective effect of PCV13 against the outcome of vaccine-type pneumococcal pneumonia in adults. The results are not anticipated to be available until at least 2012, so until then, physicians will need to base their decisions regarding pneumococcal vaccination on the limited data available.

When discussing the possible use of PCV13 in adults, an important factor to consider is that PCV13 covers 10 fewer serotypes of S pneumoniae than the PPV23. Because of the serotype replacement seen after the introduction of PCV7, strains that are not included in the PCV7 are now more frequently causing disease.²⁷ Although some of these emerging strains are included in the PCV13 (such as the increasingly common 19A strain), not all are. The most recent data from the United States demonstrate that approximately 15% of invasive pneumococcal disease is due to the 10 strains included in the PPV23 but not the PCV13.²⁷ Furthermore, once children begin receiving PCV13, we may see further serotype replacement, increasing the frequency of disease due to strains not included in PCV13. The protection that PPV23 provides for these strains, although limited, must be balanced against the possible improved protection against a smaller number of strains that is offered by the PCV13.

We believe that the emerging data regarding PCVs and the current availability of PCV13 for children creates a need for providers to be fully informed about the possible benefits of PCVs in adults. Rather than continuing to use PPV23 in all adults, some providers may consider an alternative strategy for some patients, such as using the PCV13 despite lack of Food and Drug Administration approval, perhaps with subsequent provision of PPV23. This would be similar to recommendations for use of Hib conjugate vaccine in HIV-infected adults,⁴⁰ even though the vaccine is only approved for children. Finally, if the above option is not feasible, because PPV23 may impair the response to PCV13, it is reasonable to consider the possibility that some patients would be better served by delaying pneumococcal vaccination until the PCV13 is approved for use in adults. Although the limited availability of outcome data precludes a firm recommendation, well-informed clinicians might select different options depending on a patient’s life expectancy, comorbidities, and risk for pneumococcal disease. However, unless official guidelines address potential advantages of alternative strategies, physicians may be reluctant to consider these alternative strategies because of concern regarding liability.

At least one study suggests that many physicians are not aware of the limitations of the PPV23.⁴¹ It is important that physicians and patients be given guidance about the considerations raised in this article. We have been impressed by the extremely prompt and useful guidance from various advisory groups and governmental agencies in response to the rapidly changing knowledge about the novel 2009 influenza A(H1N1) pandemic, even when the knowledge was incomplete. However, every year in the United States, many more individuals die of pneumococcal disease than have been killed by the 2009 influenza A(H1N1) pandemic.^42,43 Physicians and patients should be satisfied with no less of a sense of urgency about potential missed opportunities to prevent morbidity and death from pneumococcal disease. Although reassessing policy regarding pneumococcal vaccination will require careful analysis of available information, we encourage advisory groups to tackle this issue at the earliest opportunity.

Financial/nonfinancial disclosures: The authors have reported to CHEST the following conflicts of interest: Dr Metersky has served as a consultant and on advisory boards for Wyeth and Pfizer, and as a speaker for Merck. Dr Dransfield has served on the speaker’s bureau as a consultant for GlaxoSmithKline (< $10,000 in 2009) and served on the speaker’s bureau and as a consultant for Boehringer Ingelheim (< $5,000 in 2009). He has performed contracted COPD-related research for: Boehringer Ingelheim, GlaxoSmithKline, Pfizer, Roche, Aeris, AstraZeneca, Novartis, and Altana/Nycomed and has received grants from the National Heart, Lung, and Blood Institute for COPD-related research. Dr Jackson has served as a consultant and on an advisory board for Wyeth and has received research funding from Wyeth. She has also served as a consultant to Merck.