The effect of pneumococcal immunization on total and antigen-specific B cells in patients with severe chronic kidney disease

Abstract

Background: While the 23-valent pneumococcal polysaccharide vaccine (PPV23) is routinely used in Canada and some other countries to prevent pneumococcal infection in adults with chronic kidney disease (CKD), patients develop a suboptimal antibody response to PPV23 due to their immune dysfunction. The 13-valent pneumococcal conjugate vaccine (PCV13) has superior immunogenicity in some categories of immunocompromised adults; however, its effect on the immune response in CKD patients has only been addressed by two recent studies with conflicting results. The effect of PPV23 or PCV13 on B cells in these patients has not been previously studied. We studied the absolute numbers and proportions of B cells and subpopulations in two groups of adult patients with severe CKD pre- and 7 days post-immunization with PCV13: pneumococcal vaccine naïve and previously immunized with PPV23 (over one year ago).

Results: PPV23 immunized patients had significantly lower proportions and absolute numbers of class switched memory (CD19 + CD27 + IgM-), as well as lower absolute numbers of IgM memory (CD19 + CD27 + IgM+) and class switched B cells (CD19 + CD27-IgM-) compared to PPV23 naïve patients. Following PCV13 immunization, the differences in absolute numbers of B-cell subpopulations between groups remained significant. The PPV23 immunized group had higher proportions of CD5- B cells along with lower proportions and absolute numbers of CD5+ B cells compared to PPV23 naïve patients both pre- and post-immunization with PCV13. However, previous PPV23 immunization did not have a noticeable effect on the numbers of total IgG or serotype 6B and 14 specific antibody-secreting cells detected 7 days post-immunization with PCV13. Nevertheless, fold increase in anti-serotype 14 IgG concentrations 28 days post-PCV13 was greater in PPV23 naïve than in previously immunized patients.

Conclusions: The results suggest that immunization with PPV23 may result in long-term changes in B-cell subpopulations such as increased prevalence of CD5- B cells and decreased prevalence of class switched memory B cells in the peripheral blood. Because previous immunization with PPV23 in patients with CKD is associated with a significant decrease in the total class switched memory B cells in response to subsequent immunization with PCV13, this may reduce PCV13 immunogenicity in the setting of PPV23 followed by PCV13.

Trial registration: Registered February 24, 2015 at ClinicalTrials.gov (NCT02370069).

Keywords: 13-valent pneumococcal conjugate vaccine (PCV13); 23-valent pneumococcal polysaccharide vaccine (PPV23); B cells; Chronic kidney disease (CKD); Enzyme-linked immunospot assay (ELISPOT); Flow cytometry; Memory B cells; Streptococcus pneumoniae; T-cell dependent response; T-cell independent response.

Fig. 1

a. B-cell subpopulations in pneumococcal vaccine naïve (n = 14) and previously immunized with 23-valent pneumococcal polysaccharide vaccine (PPV23) (n = 19) patients with severe chronic kidney disease prior to immunization with 13-valent pneumococcal conjugate vaccine. Isolated peripheral blood mononuclear cells were stained for CD19, CD27, and IgM and analyzed by flow cytometry. The geometric mean of each subpopulation proportion is displayed. Statistical significance determined by Student’s t-test (*p < 0.05). b. Absolute numbers of B-cell subpopulations in pneumococcal vaccine naïve (n = 14) and previously immunized with 23-valent pneumococcal polysaccharide vaccine (PPV23) (n = 19) patients with severe chronic kidney disease prior to immunization with 13-valent pneumococcal conjugate vaccine. Isolated peripheral blood mononuclear cells were stained for CD19, CD27, and IgM. Geometric means of absolute numbers of B-cell subpopulations were determined by multiplying the proportions of B-cell subpopulations by the absolute number of lymphocytes. Statistical significance *p < 0.05 (Student’s t-test for IgM memory and class switched B cells, Mann-Whitney U test for class switched memory B cells)

Fig. 2

a. B-cell subpopulations in pneumococcal vaccine naïve (n = 25) and previously immunized with 23-valent pneumococcal polysaccharide vaccine (PPV23) (n = 35) patients with severe chronic kidney disease on day 7 post-immunization with 13-valent pneumococcal conjugate vaccine. Isolated peripheral blood mononuclear cells were stained for CD19, CD27, and IgM and analyzed by flow cytometry. The geometric mean of each subpopulation proportion is displayed. No statistically significant difference was detected between the groups. b. Absolute numbers of B-cell subpopulations in pneumococcal vaccine naïve (n = 14) and previously immunized with 23-valent pneumococcal polysaccharide vaccine (PPV23) (n = 19) patients with severe chronic kidney disease 7 days post-immunization with 13-valent pneumococcal conjugate vaccine. Isolated peripheral blood mononuclear cells were stained for CD19, CD27, and IgM. Geometric mean absolute numbers of B-cell subpopulations were determined by multiplying the proportions of B-cell subpopulations by the absolute number of lymphocytes. Statistical significance *p < 0.05 (Student’s t-test for class switched B cells, Mann-Whitney U test for IgM memory and class switched memory B cells)

Fig. 3

a. CD5+ and CD5- B-cell subpopulations pre-immunization (day 0) and post-immunization (day 7) with 13-valent pneumococcal conjugate vaccine in patients with severe chronic kidney disease. Peripheral blood mononuclear cells were stained for CD19 and CD5 and analyzed by flow cytometry. The geometric means are displayed. CD5+ and CD5- B-cell subpopulations are compared between pneumococcal vaccine naïve (day 0, n = 14; day 7, n = 15) and previously immunized with 23-valent pneumococcal polysaccharide vaccine (PPV23) (day 0, n = 19; day 7, n = 24) patients. Statistical significance *p < 0.05 (Mann-Whiney U test for day 0 CD5+, CD5- and day 7 CD5- B cells, Student’s t-test for day 7 CD5+ B cells). b. Absolute numbers of CD5+ and CD5- B-cell subpopulations pre-immunization (day 0) and post-immunization (day 7) with 13-valent pneumococcal conjugate vaccine in patients with severe chronic kidney disease. Absolute numbers of B-cell subpopulations were determined by multiplying the proportions of B-cell subpopulations by the absolute number of lymphocytes. The geometric means are displayed. Comparison between pneumococcal vaccine naïve (day 0, n = 14; day 7, n = 15) and previously immunized with 23-valent pneumococcal polysaccharide vaccine (PPV23) (day 0, n = 19; day 7, n = 24) patients is displayed. Statistical significance *p < 0.05 (Student’s t-test for day 0 CD5+ B cells, Mann-Whitney U test for day 7 CD5+ B cells)

Fig. 4

a. Numbers of total IgG antibody secreting cells (ASC) per 200,000 peripheral blood mononuclear cells (PBMC) pneumococcal vaccine naïve (n = 18) and previously immunized with 23-valent pneumococcal polysaccharide vaccine (PPV23) (n = 29) patients with severe chronic kidney disease on day 7 post-immunization with PCV13. The median is displayed counts above the upper limit of detection (200) were assigned a value of 210 (8 pneumococcal vaccine naïve and 13 previously immunized with PPV23). If no spots were detected, a value of 0.5 was assigned for statistical purposes. PPV23 naïve and PPV23 immunized each had 4 values below the limit of detection. Control wells coated with methylated human serum albumin were not displayed (all values were below the limit of detection). b. Numbers of IgG antibody secreting cells (ASC) specific for pneumococcal capsular polysaccharides of serotypes 6B or 14 per 200,000 peripheral blood mononuclear cell (PBMC) in pneumococcal vaccine naïve (n = 18) and previously immunized with 23-valent pneumococcal polysaccharide vaccine (PPV23) (n = 31) patients with severe chronic kidney disease 7 days post-immunization with PCV13 (median is shown). If no spots were detected a value of 0.5 was assigned for statistical purposes. For serotype 6B, 9 pneumococcal vaccine naïve and 16 previously immunized with PPV23 patients had ASC below the lower limit of detection. For serotype 14, 7 pneumococcal vaccine naïve and 13 previously immunized with PPV23 patients had ASC below the lower limit of detection. Control wells coated with methylated human serum albumin were not displayed (all below the limit of detection)