*3.1. Anti-Prevenar Antibodies*

Prevenar-specific IgG was significantly increased in all subjects who received any active treatment compared to placebo-treated subjects, see Table 2 and Figure 2A. Prevenar-specific IgM was significantly increased in subjects who received any active treatment up to 28 weeks, however after 68 weeks only subjects receiving three active doses had a significantly increased IgM level compared to placebo.


**Table 2.** Estimated differences for Prevenar-specific IgG and IgM levels. was significantly increased in subjects who received any active treatment up to 28 weeks, however after 68 weeks only subjects receiving three active doses had a significantly increased IgM level

*Biology* **2020**, *9*, x 5 of 10

treatment compared to placebo-treated subjects, see Table 2 and Figure 2A. Prevenar-specific IgM

ED = estimated difference, A = active, P = placebo, CI = confidence interval. 28–68 weeks (APP vs. PPP) 17,116.7 11,720.9–22,512.5 <.0001 2233.9 −4392.3–8860.1 0.4862 ED = estimated difference, A = active, P = placebo, CI = confidence interval.

**Figure** *2.* Anti-Prevenar (quantified by antibodies against pneumococcal wall saccharide) (**A**); antiphosphorylcholine (**B**); and anti–oxLDL responses (**C**). Mean + SD; *n* = 4 per group, *n* = 3 for AAP **Figure 2.** Anti-Prevenar (quantified by antibodies against pneumococcal wall saccharide) (**A**); anti-phosphorylcholine (**B**); and anti–oxLDL responses (**C**). Mean + SD; *n* = 4 per group, *n* = 3 for AAP (active-active-placebo) group and *n* = 8 for PPP (placebo-placebo-placebo) group. RLU/100ms: relative light units/100 ms 'A': active treatment, 'P': placebo treatment. Dotted lines indicate vaccination times (baseline, 4 weeks, 28 weeks). Statistical analysis using a mixed model analysis of variance with fixed factors: treatment group and time, and the interaction of treatment group and subject as random factor. For *p*-values for Prevenar responses, see Table 2. No significant changes were found for oxLDL and PC responses, except for PC-specific IgM in the APA (active-placebo-active) group compared to PPP (*p* = 0.005), and oxLDL-specific IgM and IgG responses at 68 weeks in the APA group compared to PPP (*p* = 0.007 for IgG, *p* = 0.005 for IgM).

#### *3.2. Anti-OxLDL and Anti-PC Antibodies 3.2. Anti-OxLDL and Anti-PC Antibodies*

No difference was observed in PC-specific IgG levels compared to placebo, for any of the active treatment groups. Similarly, no difference was observed for PC-specific IgM levels between active treatment groups and placebo, with the exception of IgM levels being higher in the APA group compared to PPP during the study period (ED: 9409.7, 95% CI: 3227.5–15,591.9, *p* = 0.005) (Figure 2B). No difference was observed in PC-specific IgG levels compared to placebo, for any of the active treatment groups. Similarly, no difference was observed for PC-specific IgM levels between active treatment groups and placebo, with the exception of IgM levels being higher in the APA group compared to PPP during the study period (ED: 9409.7, 95% CI: 3227.5–15,591.9, *p* = 0.005) (Figure 2B).

*Biology* **2020**, *9*, x 6 of 10

(active-active-placebo) group and *n* = 8 for PPP (placebo-placebo-placebo) group. RLU/100ms: relative light units/100 ms 'A': active treatment, 'P': placebo treatment. Dotted lines indicate vaccination times (baseline, 4 weeks, 28 weeks). Statistical analysis using a mixed model analysis of variance with fixed factors: treatment group and time, and the interaction of treatment group and subject as random factor. For *p*-values for Prevenar responses, see Table 2. No significant changes were found for oxLDL and PC responses, except for PC-specific IgM in the APA (active-placebo-active) group compared to

There were no differences in oxLDL-specific IgG and IgM antibodies between active and placebo treated subjects up to 28 weeks. However, at 68 weeks, subjects who received an active treatment at baseline and after 28 weeks (APA) had an increased oxLDL-specific IgG level compared to subjects receiving three placebo injections (PPP) with an estimated difference (ED) of 9913 (95% CI: 3141–16,686; *p*=0.007). As shown in Figure 2C difference between these groups were also observed for oxLDL-specific IgM levels (ED: 12235, 95% CI: 4179–20,290; *p* = 0.005). There were no differences in oxLDL-specific IgG and IgM antibodies between active and placebo treated subjects up to 28 weeks. However, at 68 weeks, subjects who received an active treatment at baseline and after 28 weeks (APA) had an increased oxLDL-specific IgG level compared to subjects receiving three placebo injections (PPP) with an estimated difference (ED) of 9913 (95% CI: 3141–16,686; *p* = 0.007). As shown in Figure 2C difference between these groups were also observed for oxLDL-specific IgM levels (ED: 12235, 95% CI: 4179–20,290; *p* = 0.005).

#### *3.3. Lipids 3.3. Lipids*

The levels of total cholesterol (**A**), LDL (**B**), HDL (**C**) and triglycerides (**D**) of all groups during the study are depicted in Figure 3. No significant differences were found between treatment groups, with the exception of subjects receiving a single active treatment at the beginning of the study (APP) who had a significant higher triglyceride level compared to placebo (PPP) (ED: 7.9%, 95 %CI 18%–171%; *p* = 0.009). The levels of total cholesterol (**A**), LDL (**B**), HDL (**C**) and triglycerides (**D**) of all groups during the study are depicted in Figure 3. No significant differences were found between treatment groups, with the exception of subjects receiving a single active treatment at the beginning of the study (APP) who had a significant higher triglyceride level compared to placebo (PPP) (ED: 7.9%, 95%CI 18%– 171%; *p* = 0.009).

**Figure 3.** Cholesterol (**A**), LDL (low density lipoproteins) (**B**), HDL (high density lipoproteins) (**C**) and triglycerides plasma levels (**D**) (mean + SD). *n* = 4 per group, *n* = 3 for AAP group and *n* = 8 for PPP group. 'A': active treatment, 'P': placebo treatment. Dotted lines indicate vaccinations (baseline, 4 weeks, 28 weeks). Statistical analysis using a mixed model analysis of variance with fixed factors: treatment group and time, and the interaction of treatment group and subject as random factor. No significant differences were found, with the exception of APP versus PPP for triglycerides (*p* = 0.009).

#### **4. Discussion**

The present study evaluated the effect of a 13-valent conjugate pneumococcal vaccine on the induction of anti-oxLDL and anti-PC antibodies and cholesterol levels in humans. Several vaccination regimens were tested, where subjects received either one, two or three doses of Prevenar-13 over a period of 28 weeks compared to placebo. Despite the induction of an adequate anti-Prevenar 13 antibody response, there was no evident induction of either PC-specific or oxLDL-specific antibodies. Prevenar-13 immunization induced a significant IgG2 response when subjects were immunized at least twice, while levels of IgG3 and IgG4 were not altered (Figure S1). In one active treatment group a statistically significant difference in PC-specific and oxLDL-specific antibody levels was observed compared to placebo. This occurred in the group receiving two doses of the vaccine at the start of the study and after 28 weeks (APA). A significant increase in PC-specific IgM and oxLDL-specific IgM and IgG was observed at 68 weeks. Interestingly, we observed a significant increase in PC-specific IgG3 for the group receiving two vaccinations four weeks apart (AAP) compared to placebo (PPP) (Figure S1 panel B). Elevated oxLDL-specific IgM is believed to be atheroprotective [10–13], but the role of oxLDL-specific IgG levels is not fully elucidated. Laczik et al. [16] showed that increased oxLDL-specific IgG levels correlate with acute coronary syndrome, while immunization with oxLDL, resulting in an increase in oxLDL-specific IgG, resulted in decreased plaque development in several mouse models [18]. Furthermore, oxLDL levels are inversely correlated with oxLDL-specific IgG serum levels [17]. Binder et al. [19] showed *S. pneumoniae* immunization in mice induced a much stronger oxLDL-specific IgM response than an oxLDL-specific IgG response (100,000 vs. 10,000 RLU/100ms). The current clinical data are at odds with this observation, though the observed anti-oxLDL responses in the APA group may indicate that the timing of vaccination could be important.

Previous studies have investigated the 23-valent polysaccharide pneumococcal vaccine as a means to elicit oxLDL-specific antibodies in humans, with conflicting results [26,27,29]. One study reported that, after vaccination with the 23-valent vaccine, an increased oxLDL-specific IgG antibody titer was observed compared to healthy, unvaccinated individuals (248 U/l vs. 55 U/l). An effect of vaccination on IgM was not reported. In the other two studies, no association between vaccination status and oxLDL-specific antibodies was observed. There are, however, key differences between these studies and the present study. First, patients in all three referenced studies only got a single vaccination, whereas in the present study, up to three vaccinations were given. Second, there is a major difference between the 23-valent polysaccharide vaccine that was used in these studies and the 13-valent polysaccharide conjugate vaccine that was used in the present study. The latter is constituted of cell wall polysaccharides that have been conjugated to a protein. The 13-valent vaccine is thereby considered to be more powerful in eliciting an antibody response against *Streptococcus* pneumonia [30,31]. These data were the basis for the selection of the 13-valent vaccine for the current clinical study.

Although the sample size per group was relatively small in the current study (*n* = 4 for treatment groups, *n* = 8 for placebo), the study was sufficiently powered to detect Prevenar-13-induced rises in IgM titers, had these occurred as in the murine experiments [19].

The induction of anti-pneumococcal wall saccharide antibodies demonstrates that the 13-valent vaccine was effective for its intended use. However, the vaccine did not elicit a robust oxLDL-specific IgM response, as observed in mouse experiments. One explanation for the poor induction of oxLDL-specific antibody responses by Prevenar 13 may be that the murine immune response is poorly translatable to humans. Human and mice have numerous discrepancies in their innate and adaptive immune systems, such as cytokine receptor and costimulatory molecule expression and function [32,33]. Moreover, murine studies are commonly performed in inbred strains, with limited genetic variability between mice. This, and the fact that mice are kept in a more-sterile environment, results in a smaller immune diversity compared to humans [34]. On the other hand, the mild but significant induction of oxLDL-specific antibody responses in one active treatment group (APA) does support further clinical investigation of mimicry between pneumococcal vaccination and oxLDL. Furthermore, it could be

hypothesized that the PC content in the Prevenar-13 vaccine used was not high enough to induce robust PC- and oxLDL-specific responses.
