Specific antibody deficiency is characterized by an abnormal IgG antibody response to a pneumococcal vaccine, which was developed to protect against the Gram-positive cocci S. pneumoniae (12). The Centers for Disease Control and Prevention recommend 13-valent pneumococcal conjugate vaccine (PCV13) and 23-valent pneumococcal polysaccharide vaccine (PPSV23) for adults, and a series of PCV13 vaccinations for children under 2 years of age (28, 29). Vaccine indications and timings depend on age, previous vaccinations, and the presence of high-risk conditions including congenital immunodeficiency [see Figure 1 and Ref. (28) for further information].

Response to pneumococcal vaccines is usually determined by assessing levels of IgG specific for serotypes included in the vaccine by multiplex bead immunoassay or enzyme-linked immunosorbent assay (30–32). Radioimmunoassay has also been used in the past, although this method does not differentiate between antibody responses by different Ig isotypes (i.e., those other than IgG) (33). To date, the diagnosis of SAD has been hindered by a lack of controlled clinical studies and the absence of a standardized definition of an insufficient pneumococcal polysaccharide antibody response. Furthermore, and as alluded to above, the criteria for diagnosis as established by consensus groups have changed over the years.

Specific antibody deficiency is normally diagnosed by determining the ability to generate protective titers in response to pneumococcal vaccines (12); however, it is important to note that the definition of a protective titer is not uniform and may vary depending on the nature of the vaccine (12, 34, 35). A serotype-specific level of 1.3 μg/mL has been considered protective with respect to invasive disease following polysaccharide immunization (35, 36), and other studies have shown that levels of 0.35 µg/mL were deemed to provide protection against invasive pneumococcal infections following immunization with a conjugate pneumococcal vaccine (37). However, these studies are based on small cohorts and protective levels in response to pneumococcal vaccination and should be interpreted with caution (38). Furthermore, the level of specific antibody necessary to provide protection against infection in spaces such as the sinuses and middle ear has not been established.

An adequate response to immunization was previously defined as at least a fourfold increase in antibody levels over baseline for a given serotype (34, 39). However, the fourfold response criteria is no longer preferred as there is evidence to suggest that subjects with high baseline titers may not develop such an increase following vaccination (40). In a meta-analysis of antipneumococcal antibody responses in healthy individuals, the ratios of pre- to post-vaccination titers varied widely and depended on the particular serotype and the baseline level of antibody (36). One current recommendation is to accept a twofold response if the baseline level is ≥1.3 μg/mL (35), although the authors recommend that these data should be interpreted along with clinical correlation.

Controversy also exists over the precise number of pneumococcal serotypes needed to determine a normal response. This problem is compounded by the different pneumococcal vaccines that have been used historically. PPSV23 protects against 23 capsular serotypes (Table 2) and is a pure polysaccharide vaccine, meaning that it induces a T-cell-independent response by stimulating B-cells in the absence of T-helper cells (29). For this reason, polysaccharide vaccines are not reliably immunogenic in children under 2 years of age; thus, pneumococcal conjugate vaccines were developed, which generate a T-cell-dependent antibody response and are effective in children under 2 years (29). Earlier conjugate preparations contain 7 serotypes, while newer ones contain 13 (41). The availability of different vaccines with different antigens makes it difficult to standardize responses; for example, a retrospective report showed that PCV non-responders, including patients with SAD, may have good clinical and serological responses to PPSV23 (42).

Furthermore, there has never been a study evaluating the correlation between degree of responsiveness and infection susceptibility. A working group report on diagnostic vaccination in PIDD recommend that a normal response to pneumococcal vaccines is a response to ≥50% of serotypes for patients under 6 years of age and a response to ≥70% of serotypes for patients over 6 years of age (35). In support of these specified thresholds, the meta-analysis of pneumococcal responses in healthy individuals showed that the majority of subjects could mount at least a twofold response to most serotypes (36). Recent attention has been focused upon other studies that report an adequate response as ≥1.3 μg/mL for >50% of serotypes (19, 40).

Although controversies exist regarding the definition of a protective titer, guidelines from a working group report were developed using the best evidence currently available to describe the diagnosis of mild, moderate, severe, and memory phenotypes of deficient response, based on response to PPSV23 (Table 3) (35). Patients with a mild phenotype have multiple serotypes to which they did not generate protective titers or were unable to increase titers twofold. Patients with a moderate phenotype produce protective titers to three or more serotypes but to <50% of serotypes for those under 6 years of age or <70% of serotypes for those over 6 years of age. A severe phenotype is described as producing protective titers against two or fewer serotypes, and those protective titers generated tend to be low. Patients with a memory phenotype of deficient responses initially mount an adequate response to vaccination but do not sustain the response beyond 6 months. It is important to note that pure polysaccharide vaccines invoke a T-cell-independent response and as such do not generate a long-lived memory B-cell response (although they can boost them if the patient has previously received a conjugate vaccine); the term “memory phenotype” refers to patients who lose an adequate response to PPSV23 more quickly than usual.

For the diagnosis of SAD, titers against pneumococcal vaccine serotypes must be measured preimmunization and postimmunization (4 weeks after vaccination), and it is of greatest importance to consider whether the final antibody titer values are above protective limits. Fold increases are less relevant as patients with high preimmunization titers may not show a significant increase in antibody concentrations after vaccination (see above), so it may make sense not to include the serotype of a very high pre-vaccination titer in the post-vaccination analysis. Moreover, fold increases in patients with very low titers may be irrelevant (12). When considering the response to PPSV23, it is also imperative to take into account vaccination history. The practice parameter from the American Academy of Allergy, Asthma and Immunology (AAAAI), American College of Allergy, Asthma and Immunology (ACAAI), and Joint Council of Allergy, Asthma and Immunology (JCAAI) recommends that for patients who have previously received at least one dose of conjugate vaccine, normal antibody levels against serotypes in the conjugate vaccine do not exclude the diagnosis of SAD; thus, at least six serotypes should be tested that are present in PPSV23 only (12). Finally, without substantive and much needed natural history studies, it is also important to consider that SAD may be a transient issue in some patients, especially children, and may resolve over time (43).

Age has an important influence on the level of response to most pneumococcal polysaccharide antigens. The diagnosis of SAD should not be considered in patients less than 2 years of age where there is an overlap between the characteristic findings of SAD and those of healthy children. The responses to polysaccharide antigens, in particular, are less reliable in children under 2 years of age (13, 44). It has been demonstrated that some serotypes do not elicit a significant antibody response in small children, whereas others can elicit a high response; in one study, 78% of 1-year-old children had good antibody responses to >50% of serotypes in the pneumococcal polysaccharide vaccine (14). Another study has shown that children aged under 1 year were unable to mount an antibody response to serotype 14, but this serotype elicited the highest response in elderly individuals (45). Inversely, children aged 1 year mounted high antibody responses to serotype 3, whereas adult individuals were only able to mount weak responses (45). Thus, anti-polysaccharide antibody responses are age dependent and also likely to be serotype dependent. As further research is clearly needed in this area, patients cannot be reliably diagnosed based on polysaccharide vaccination response until they are over 2 years old and their immune system has matured sufficiently.

It is extremely challenging to standardize the diagnostic approach for patients in whom SAD is suspected. The sentinel pathogen in patients with SAD is considered to be S. pneumoniae although its precise role in causing disease in these patients is not established. Typical infections include otitis media, sinusitis, bronchitis, and pneumonia (41); however, the objective documentation of upper and lower respiratory tract infections outside of pneumonia is poor. Sepsis, meningitis, and osteomyelitis may also occur, though less commonly. It is assumed in most cases, particularly in less clinically severe infections, that pneumococcus has a prominent role, although definitive microbiologic proof of its presence or causal relation to disease is rarely sought and often difficult to demonstrate. Culture data may be available for more invasive infections. The presence and pathogenicity of other common respiratory flora such as H. influenzae, Moraxella catarrhalis, etc. have not been studied, but are to some extent assumed along with the role for pneumococcus by extrapolation from studies of patients with antibody deficiencies who have similar presentations together with overt hypogammaglobulinemia (46). Contributions of viral agents to symptoms may also be assumed, but this also has not been formally studied. It should be noted that evidence from some small studies indicates that memory switched B-cell percentage is a good indicator of clinical complications associated with SAD (and also CVID) (23).

In any evaluation of patients for immunodeficiency, infections should be documented in as much detail as possible with appropriate culture and imaging data, and documentation of response to therapy. Other factors can come into play that may increase the frequency and severity of upper respiratory tract infections including smoking, daycare attendance, and atopic disease; these factors should be considered and treated or abated.

For patients who have best objective criteria for recurrent upper and/or lower respiratory tract infections, or those who have had a serious documented infection with pneumococcus or other bacterial pathogens (i.e., bacteremia) for which they have been vaccinated, consideration of the SAD diagnosis in those over 2 years of age is appropriate. Patients should have a standard screening immunological assessment consistent with the practice parameter to include quantitative Ig levels and titers (12). Assuming that other immunodeficiencies are not suspected based upon the laboratory tests obtained and the IgG level is normal, but antibody titers to pneumococcal polysaccharides are absent or low, then unconjugated pneumococcal vaccination should be administered (Figure 1). Titers should be measured at approximately 4 weeks thereafter. If the response is adequate, titers should be measured again 6 months after the vaccination in consideration of a possible “memory phenotype” of deficient vaccination response, or later if infections recur after initial improvement. If this assessment does not show substantive decreases in titer, the diagnosis of SAD should not be considered further. If the 4-week post-vaccination testing is abnormal, intervention can be considered based upon the severity of the hyporesponsiveness and clinical presentation.

There is little evidence to guide the use of antibiotic prophylaxis in patients with SAD, or indeed in patients with PIDDs, and current practice is not based on data from patients specifically diagnosed with SAD, but from immune competent patients with recurrent acute otitis media, chronic rhinosinusitis, cystic fibrosis, and bronchiectasis (24). For example, in 1 prospective double-blind study, 24 children were identified with bacterial respiratory infections that continued over a 4-month observation period. All of the 7 children treated with placebo continued to have bacterial respiratory infections, whereas 14 of 16 children treated with trimethoprim–sulfamethoxazole (TMP–SMX) became infection-free (p = 0.002) (47). Although the majority of experts use antibiotic prophylaxis in practice (48, 49), studies are needed to determine the optimal dose, duration, and choice of antibiotic. The topic of prophylactic antibiotics has recently been reviewed by Kuruvilla and de la Morena; however, SAD was not included in this discussion (50).

In practice, both antibiotic prophylaxis (and IgG) therapy are widely used to treat SAD. A survey of PIDD management of 405 allergists and immunologists in the United States was performed, and the findings were reported separately for general and specialized immunologists (based on patients with PIDD comprising more than 10% of their clinical practice) (49). The majority reported using antibiotic prophylaxis in patients with SAD, and there was no significant difference in the percentage of general and specialized immunologists using antibiotic prophylaxis. While less than 25% of general immunologists reported that this treatment was moderately or extremely useful, more than twice this number of specialized immunologists considered antibiotic prophylaxis useful and of value. We concur with this perspective and support the use of antibiotic prophylaxis as the first-line therapy where pneumococcal conjugate vaccination fails to provide protection. There is no consensus protocol regarding the frequency and severity of infections that should motivate clinicians to begin antibiotic prophylaxis; however, patients with two or more episodes of pneumonia in 1 year or with multiple (>4–5) episodes of otitis media or sinusitis could be considered for treatment. While the individual regimens applied to patients vary substantially, the AAAAI, ACAAI, and JCAAI practice parameter recommends regimens including azithromycin (500 mg weekly or 250 mg every other day in adults; 10 mg/kg weekly or 5 mg/kg every other day in children) and TMP–SMX (160 mg daily or twice daily in adults; 5 mg/kg daily or twice daily in children) (12). Some patients may require year-round prophylaxis; however, in patients who have seasonal variation in their susceptibility to infection, we have considered the seasonal application of antibiotic prophylaxis. We are unable to offer specific guidance as to duration of antibiotic prophylaxis, but patients should be evaluated every 6 months at least to assess interval history, infections, treatment and response, etc. Those who continue to have infections on antibiotic prophylaxis, or who cannot tolerate long-term antibiotics, should be considered as candidates for IgG supplementation.

A number of published expert guidelines and opinions have recommended IgG therapy in patients with SAD (12, 24, 40, 51–53), as well as observations from retrospective studies (20, 54–56), although definitive data are lacking. The practice parameter from the AAAAI, ACAAI, and JCAAI recommends that some patients with SAD might benefit from a period of IgG replacement therapy, and that such a determination can be based on immunological and clinical severity, and unresponsiveness or adverse effects of other medical interventions (12). Other parameters outlined by the UK Department of Health recommend that IgG therapy should be given only if antibiotic therapy is ineffective and impaired antibody production is demonstrated (52, 53), which is consistent with opinions from other experts (24, 40, 53). Recommended doses of IgG for treatment of SAD are the same as those used to treat other PIDDs and are based on clinical outcome (52, 53); however, there are no evidence-based criteria to determine the optimal duration of IgG replacement therapy. The AAAAI, ACAAI, and JCAAI practice parameter recommends that young patients who have stabilized after a period of IgG treatment, and are deemed at low risk for relapse, should discontinue IgG treatment for a period of 4–6 months to re-evaluate therapy (12). Our experience suggests an initial IgG treatment duration of 6–12 months, depending on clinical circumstance, in which a patient has not taken prophylactic antibiotics is a reasonable approach. Following this, a consideration to stop IgG therapy may be made after an extended period of time, for example, 1–2 years, if the patient has been without infection and required no or very few courses of antibiotics. Patients with a poor response to prophylactic antibiotics should be reassessed after 6 months. Also, in some (but not all) patients, IgA and IgM levels, total B-cell numbers, and proportions of memory B-cells may also normalize over time and be used as a possible indicator of improvement in humoral immune function. It should be considered that spontaneous resolution of humoral immunodeficiency in children is more common than in adults.

The effectiveness of IgG therapy in preventing infections in patients with SAD has been assessed in retrospective studies. In an observational study of 91 children with recurrent respiratory infections, 10 patients with SAD were identified who did not respond to pneumococcal vaccine. IgG therapy was initiated in two of these children who had inadequate responses to four of the seven serotypes tested. Prior to diagnosis, 1 patient had experienced 25 episodes of acute otitis media and the other patient had suffered from 5 episodes of pneumonia and 3 of acute otitis media. Both children remained healthy during and after IgG treatment (55). In a retrospective study of 75 patients, 30 received 400 mg/kg/month IgG therapy as prophylaxis against recurrent infections that had continued despite antibiotic therapy and treatment of concomitant allergic diseases (54). Patients with fewer antibody responses to pneumococcal vaccination were more likely to require IgG therapy (p < 0.01), and the number of infections was significantly reduced following IgG therapy (p < 0.001). In another retrospective analysis, 20 patients with difficult-to-treat asthma were subsequently diagnosed with SAD. A dose of 400–600 mg/kg IgG administered intravenously every 3–4 weeks reduced morbidity, hospitalizations, and respiratory infections (56). In another retrospective chart review of 239 adults with chronic rhinosinusitis, 56 (23%) were diagnosed with SAD (20). Of these patients, 10 were treated with IgG replacement therapy. All 10 patients had fewer infections following IgG replacement therapy, and 7 were deemed to have “improved greatly.” Further studies are needed to evaluate the efficacy of IgG therapy in patients with SAD and to optimize treatment strategies.

The majority of immunologists in the US-based PIDD management survey mentioned above also reported using IgG therapy to treat at least some patients with SAD, and there was no significant difference in IgG use between specialized and general immunologists. A similar survey was also performed with experts from the AAAAI and also ESID (48). Similar to the previous survey, the results showed that approximately half of all immunologists recommended IgG therapy for at least 5–50% of patients with SAD and there was no significant difference between the percentages of specialized and general immunologists, nor in the percentages of experts from the AAAAI and ESID, who prescribe IgG therapy. These results indicate that although there is a lack of guidance regarding IgG use in patients with SAD, use does not appear to vary significantly by region or depending on the specialty of the immunologist.

In summary, although there is a lack of uniform recommendations and clear guidance for the treatment of SAD, we believe that it is most advisable to consider IgG therapy in those who have some combination of the following features: (a) severe or very frequent recurrent infections; (b) poor response to pneumococcal polysaccharide vaccination; (c) inability to tolerate antibiotic prophylaxis due to multiple hypersensitivity, severe side effects or complications such as Clostridium difficile colitis, etc.; or (d) failure to respond to prophylactic antibiotics.