Background
Subnormal IgG subclass (IgGSc) concentrations in some adults represent a primary immunodeficiency disorder(s) characterized by frequent or severe respiratory tract infections, suboptimal IgG responses to pneumococcal polysaccharides, and increased prevalence of autoimmune disorders and atopy [
1‐
3]. The subnormal serum concentrations are mainly present in any one or more of three main subclasses IgG1, IgG2, or IgG3 [
4,
5]. Molecules of the four respective IgG subclasses are characterized by distinct induction antigens or allergens, antigen binding, immune complex formation, complement activation, half-life, and other properties. Accordingly, patients with different subnormal IgGSc immunophenotypes have dissimilar predisposition to infection [
6].
IgG1 production is induced by exposure to soluble and membrane protein antigens and allergens. Some individuals with subnormal IgG1 have frequent or severe infections, especially of the respiratory tract [
7,
8]. Antibody responses to bacterial capsular polysaccharide antigens are associated with low serum IgG2 [
6,
9,
10]. Some persons with subnormal IgG2 have increased risk of respiratory tract infections due to encapsulated bacteria, although others are apparently healthy [
11,
12]. Some persons with subnormal IgG3 also have subnormal concentrations of other subclasses, especially IgG1 [
13,
14]. Some allergens and parasites elicit IgG4 responses [
15,
16], but the clinical significance of subnormal IgG4, if any, is unclear.
Some patients with subnormal IgGSc respond to certain vaccines and others do not [
8,
17‐
19]. In a study of 17 patients with subnormal IgGSc, 14 patients who responded to 23-valent pneumococcal polysaccharide vaccination (PPSV23) had no further progression of respiratory tract infections, whereas three PPSV23 non-responders required antibiotic treatment [
20]. van Kessel and colleagues reported that following vaccination with PPSV23, non-responders with subnormal IgG1 to the measured 14 pneumococcal polysaccharide serotypes (PPS), had increased risk of infection with
Streptococcus pneumoniae, and a greater proportion had received treatment with corticosteroids [
21]. Because IgGSc levels and vaccination response are independent but overlapping markers of B-lymphocyte and plasma cell function, it is plausible that the combination of a lack of vaccination response and subnormal IgGSc increases risk of corresponding infection(s).
In this report, we describe a retrospective analysis of characteristics of 59 adults with subnormal IgGSc concentrations but normal IgG. We report whether they did or did not respond to PPSV23 and their concentrations of pre- and post-PPSV23 serotype-specific IgG. We discuss the relationships between subnormal IgGSc immunophenotypes and other attributes of the present patients with specific and aggregate IgG responses to PPSV23 and the pertinence of the present results to infection susceptibility and prevention in patients with subnormal IgGSc.
Methods
Subject selection
Performance of this work was reviewed by the Institutional Review Board of Brookwood Medical Center, Alabama. Obtaining informed consent was not required because this study involved evaluation of observations obtained in routine medical care. We studied the records of unrelated non-Hispanic white adults (≥18 years of age) referred to a single practice in a large suburban medical center that evaluates and treats many adult patients with primary immunodeficiency. All patients presented with frequent or severe bacterial infections of the upper or lower respiratory tract and were diagnosed to have IgG subclass deficiency [
22]. By selection, all of the present patients had normal total serum IgG. Autoimmune conditions, atopy, and other allergy manifestations were defined as previously described [
23]. The present patients were evaluated during the interval 2012–2018.
Other conditions
Body mass index was computed as kg/m
2. We classified diabetes according to the criteria of the American Diabetes Association [
24].
Polyvalent pneumococcal polysaccharide vaccination
S. pneumoniae serotype-specific IgG concentrations were measured before and after PPSV23 (Pneumovax®23; Merck & Co., Inc., Kenilworth NJ, USA). The median interval between pre- and post-PPSV23 testing was 33 days (range 19–158 days).
On 30 December 2011, PCV13 (Prevnar13®, Wyeth Pharmaceuticals, Inc., Philadelphia, PA, USA) was approved for use in the US among adults aged ≥50 years to prevent pneumonia and invasive disease caused by
S. pneumoniae serotypes contained in the vaccine [
25]. On 13 August 2014, the routine use of PCV13 among adults aged ≥65 years was recommended in series with PPSV23, the vaccine then recommended for adults aged ≥65 years [
25]. In late 2018, the US Centers for Disease Control and Prevention recommended pneumococcal vaccination (both PPSV23 and PCV13) for adults ages 19–64 years who have certain medical conditions or who smoke, including those with congenital or acquired immunodeficiency. The US Centers for Disease Control and Prevention now recommends pneumococcal vaccination for all adults 65 years or older (both PPSV23 and PCV13) [
26]. Although none of the present adults reported that they had received PCV13, the possibility that some of them received PCV13 as children cannot be excluded.
Laboratory methods
All testing was performed before IgG replacement therapy was initiated. Serum Ig concentrations were measured using standard methods at a single reference laboratory (Laboratory Corporation of America, Burlington NC, USA). We defined mean ± 2 SD as reference ranges for all Ig concentrations. Ig reference ranges employed were: IgG 7.0–16.0 g/L (700–1600 mg/dL); IgG1 4.2–12.9 g/L (422–1292 mg/dL); IgG2 1.2–7.5 g/L (117–747 mg/dL); IgG3 0.4–1.3 g/L (41–129 mg/dL); IgG4 0–2.9 g/L (1–291 mg/dL); IgA 700–4000 mg/L (70–400 mg/dL); and IgM 400–2300 mg/L (40–230 mg/dL). Subnormal Ig concentrations were defined as those below the corresponding reference limits.
Pre- and post-PPSV23 serotype-specific IgG antibodies were measured by clinical laboratories (Laboratory Corporation of America, Burlington NC, USA and ViraCor-IBT, Lee’s Summit MO, USA) and included S. pneumoniae serotypes 1, 3, 4, 8, 9. 12, 14, 19, 23, 26, 51, 56, 57, and 68. Diluents for patient samples contained C-polysaccharide and polysaccharide type 22. The sums of post-PPSV23 serotype-specific IgG concentration in test panels were defined as summated concentrations.
Response to PPSV23
We defined protective serotype-specific IgG levels as > 1.3 μg/mL. We defined responders as patients who achieved protective levels of serotype-specific IgG post-PPSV23. We defined aggregate PPSV23 response as serotype-specific IgG > 1.3 μg/mL for ≥70% of serotypes tested post-PPSV23.
Statistics
The analytic data set consisted of complete observations on 59 patients with subnormal IgGSc and total serum IgG > 700 mg/dL. For some analyses, we separated patient data into three groups: 1) subnormal IgG1; 2) subnormal IgG3; and 3) subnormal IgG1 + IgG3. Descriptive data are displayed as enumerations, percentages, mean ± 1 SD, or mean 95% confidence intervals (CI). Age and BMI data were normally distributed and were compared using Student’s t-test (two-tailed). Proportions were compared using Pearson’s χ2 test or Fisher’s exact test, as appropriate. We computed odds ratios (OR) 95% CI for some observations. Means of continuous variables among three patient subgroups were compared using one-way analysis of variance (ANOVA) or Kruskal-Wallis one-way ANOVA, as appropriate. Median concentrations were compared using Mann-Whitney U Test and Spearman’s rank test to assess the association between pneumococcal summation and coverage. Dichotomous variables in frequency analyses were: male; diabetes; autoimmune condition(s); atopy (allergic asthma, allergic rhinitis, or allergic dermatitis/eczema); and other allergy (urticaria, angioedema, or anaphylaxis). Total clinical variables were calculated from these dichotomous variables.
We compared the general characteristics of patients who did or did not respond to pneumococcal polysaccharides using univariate analyses. We also performed logistic multivariable regressions on aggregate IgG response to pneumococcal polysaccharides to identify significant associations using these independent variables: age at diagnosis; male sex; body mass index (BMI); autoimmune condition(s); atopy; other allergies; subnormal IgG1; subnormal IgG3; subnormal IgG1 + IgG3; and serum levels of IgA and IgM. We computed OR of independent variables with significant associations. Dichotomous variables in frequency analyses were: male; diabetes; autoimmune condition(s); atopy (allergic asthma, allergic rhinitis, or allergic dermatitis/eczema); and other allergy (urticaria, angioedema, or anaphylaxis). Total clinical variables were calculated from these dichotomous variables. Analyses were performed with GraphPad Prism v. 5.04 (GraphPad Software, Inc., La Jolla, CA, USA), Excel 2000® (Microsoft Corp., Redmond, WA, USA), and GB-Stat® (v. 10.0, 2003, Dynamic Microsystems, Inc., Silver Spring, MD, USA). We defined values of p < 0.05 to be significant. Bonferroni corrections were applied to control the type I error rate at 0.05 for multiple univariate comparisons.
Discussion
The restriction of anti-pneumococcal polysaccharide IgG response post-PPSV23 was greater in our patients with subnormal IgG3 than in patients with subnormal IgG1. Previous studies have reported associations between subnormal IgG3 and respiratory tract infection [
19,
27]. Risk for infections due to particular
S. pneumoniae serotypes is probably increased in some patients with subnormal IgGSc due to their inability to synthesize specific protective antibodies. For example, impaired production of IgG anti-pneumococcal serotype 3 has been associated with recurrent pneumonia, empyema, septic shock, and increased fatality [
28‐
30]. There is a positive association between invasive pneumococcal infection and impaired production of anti-serotype 4, 14, 51 and 56 [
29]. IgG3 is also a key determinant in response to
Moraxella catarrhalis and
S. pyogenes infections [
31,
32].
IgG3 antibodies in commercial intravenous IgG preparations have been reported to have the most potent binding and opsonic activity for
S. pneumoniae serotype 6B [
33], a major etiology of infections in persons with primary immune deficiency [
34]. Individuals with subnormal serum IgA or heavy chain constant region deletions of the γ2 gene produce IgG1 and IgG3 anti-pneumococcal polysaccharide antibodies [
35,
36]. Some patients with recurrent bacterial respiratory tract infections and subnormal IgG3 have subnormal subsets of blood T- or B-lymphocytes or impaired lymphocyte function [
18,
23,
37,
38]. Patients with recurrent bacterial respiratory tract infections and subnormal IgG3 had less frequent or severe infections after they were treated with IgG devoid of IgG3 [
39], suggesting that qualitative deficits in IgG1, IgG2, or possibly IgG4 subclasses increase infection susceptibility. Taken together, these observations indicate that multiple factors in individuals with subnormal IgG3, including suboptimal response to some
S. pneumoniae serotype-specific antigens, contribute to increased susceptibility to respiratory tract infection.
In this study, patients with subnormal IgG3 and subnormal IgG1 + IgG3, PPSV23 responders had similar serum concentrations of anti-pneumococcal polysaccharide antibodies. In PPSV23 non-responders, the concentrations were higher in patients with subnormal IgG1 + IgG3. The greater responses to serotypes 8 and 9 and greater serum levels of IgG2 in patients with subnormal IgG1 + IgG3 could partly account for this difference.
In the cohort of 59 patients with subnormal IgGSc, there was a predominance of women, consistent with other subnormal IgGSc case series. A predominance of women was also observed in each of the three subnormal IgGSc immunophenotype subgroups we studied. In univariate analyses, we observed a strong positive association between low concentration of 1, 3, 14 and 51 serotype-specific antibodies and male gender. In a logistic regression on aggregate IgG responses, the odds of responses in men were low (0.09 [95% CI: 0.01, 0.77]). In an analysis of data extracted from 84 studies, Falagas and colleagues concluded that males in all age groups develop respiratory tract infections more frequently than females, except in cases of sinusitis, otitis externa, and probably tonsillitis [
40].
There was an inverse correlation of PPSV23 response with age in the present study. This was most evident in non-responders with subnormal IgG1 (
r = − 0.79) and responders with subnormal IgG3 (
r = 0.71). In another study, IgG, IgA, and IgM concentrations were lower in elderly adults (aged 70–79 years) than in younger adults (college students) [
41]. In another report, IgG responses to PPSV23 were stable in Prevnar13®-naive healthy adults as long as 6 years post-PPSV23 [
42]. In the current study, the serum levels of post-PPSV23
S. pneumoniae serotype-specific IgG decreased with age, suggesting that the PPSV23 response stability of adults with subnormal IgGSc may be less durable than that of adults not selected for having subnormal IgGSc.
Univariate analyses demonstrated that restricted response to certain
S. pneumoniae polysaccharide antigens was associated with increased prevalence of atopy (serotypes 4, 14, 56, and 68), other allergies (serotypes 3 and 51), and autoimmune condition(s) (serotypes 3 and 51) and with a predominance of male gender in patients with subnormal IgG3. In a regression analysis of aggregate response to PPSV23, atopy was associated with lower odds for responses, after adjustment for other variables. Atopy, defined as allergic asthma, allergic rhinitis, or allergic dermatitis/eczema, occurred in 29% of the present patients and in 36% of present patients with subnormal IgG3. Atopy and asthma were common in two case series of patients with subnormal IgG1 [
7,
8] and in 121 adult patients with subnormal IgG3 [
19]. Allergic asthma increases risk of lower respiratory tract infections in subjects not selected for subnormal immunoglobulin level [
43,
44]. In another study, viral or bacterial infections were detected in 70% of inpatients with exacerbation of asthma [
45]. Likewise, allergic rhinitis increases risk of sinusitis [
46]. These observations suggest that the combination of restricted response to certain
S. pneumoniae polysaccharide antigens and atopy increase the risk of respiratory tract infection in persons with subnormal IgGSc.
In the present adults, PPSV23 response rates decreased with increasing age. IgG responses of elderly adults to PPSV14 or PPSV23 measured by avidity were similar to those of younger adults [
47,
48]. In contrast, concentration ratios of IgG1/IgG2 produced in response to an investigational 5-valent pneumococcal vaccine declined in adults with greater age for all serotypes studied (6B, 14, 18C, 19F, and 23F) [
49]. In another study of elderly adults, post-vaccination IgG antibody concentrations for two serotypes (6B and 19F) of the five studied (4, 6B, 14, 19F, and 23F) were significantly lower and in vitro opsonophagocytosis of all five IgG serotype-specific antibodies was significantly reduced [
50]. In two other reports, serotype-specific pre- and post-immune antibody responses measured by opsonophagocytosis were also significantly lower in elderly adults [
51,
52]. IgG responses to PPS4 and PPS14 in elderly adults were associated with decreased V
H repertoire (mostly V
H3) and somatic mutations and with greater loss of oligoclonality than younger adults [
51]. Antigen responses may differentially involve more V
H4 alleles in elderly than younger adults [
53].
Aggregate IgG response was lower in the present men than women with IgGSc deficiency, after adjustment for other variables. Women have higher basal levels of plasma or serum IgG [
54,
55]. Mean serum IgG levels in 26 men with Klinefelter’s syndrome were significantly higher than those of control men, but after androgen replacement therapy declined to mean levels similar to those of controls [
56]. IgG responses to major pneumococcal serotypes in both pre- and post-PPSV23 sera were higher in elderly men than elderly women [
48,
57,
58] but decreased significantly in women with increasing age [
57]. Neither age, sex, IgG level, or IgG subclass levels were significant independent predictors of 4-fold or greater antibody response to PPSV23 [
59].
The present results demonstrate that IgGSc responses to PPVS23 in adults with IgGSc deficiency are preferential and significantly associated with pre-vaccination IgGSc immunophenotypes. In IgG concentrates used to treat patients with primary antibody deficiency, IgG2 was the most active subclass of anti-
Streptococcus pneumoniae antibodies by avidity and opsonophagocytosis measures, except that IgG3 was the most active against serotype 6B [
33]. Relatively low responsiveness to PPSV23 of the present patients with subnormal IgG3 may be due in part to their inheritance of Gm3 polymorphisms (allotypes) [
60,
61]. Two of the present patients had IgA deficiency. In another study, adults with IgG2/IgA deficiency had no IgG response to PPSV14 In normal adults, the predominant PPSV14 response occurred in IgG2 subclass [
10], although PPSV14 responses also occur in other IgG subclasses [
62]. In a subject with pneumococcal polysaccharide vaccination-specific IgG2 deficiency, the balance of specific IgG subclass antibodies changed from IgG2 > IgG1 > IgG3 > IgG4 pre-vaccination to IgG1 > IgG3 > IgG2 > IgG4 post-vaccination [
6]. In patients with IgG2 deficiency, enhancement of IgG1 and IgG3 antibody responses to pneumococcal polysaccharide antigens suggest a change in normal switching or alteration in the antigen affinity of IgG subclass-specific antibody [
36].
A strength of the present study is our evaluation of a sufficient number of adults with common subnormal IgGSc immunophenotypes to detect significant differences in some antigen-specific and aggregate IgG responses to PPSV23 and statistical associations of responses and non-responses to other variables. Limitations of the present study include exclusion of patients with subnormal IgG2 and lack of long-term follow-up of PPSV23 response stability. Investigating mechanisms by which variables such as male sex, atopy, and autoimmunity influence PPSV23 responses in patients with subnormal IgGSc was beyond the scope of the present study.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.