Peptide-based anti-PCSK9 vaccines - an approach for long-term LDLc management

Abstract

Background: Low Density Lipoprotein (LDL) hypercholesterolemia, and its associated cardiovascular diseases, are some of the leading causes of death worldwide. The ability of proprotein convertase subtilisin/kexin 9 (PCSK9) to modulate circulating LDL cholesterol (LDLc) concentrations made it a very attractive target for LDLc management. To date, the most advanced approaches for PCSK9 inhibition are monoclonal antibody (mAb) therapies. Although shown to lower LDLc significantly, mAbs face functional limitations because of their relatively short in vivo half-lives necessitating frequent administration. Here, we evaluated the long-term efficacy and safety of PCSK9-specific active vaccines in different preclinical models.

Methods and finding: PCSK9 peptide-based vaccines were successfully selected by our proprietary technology. To test their efficacy, wild-type (wt) mice, Ldlr+/- mice, and rats were immunized with highly immunogenic vaccine candidates. Vaccines induced generation of high-affine PCSK9-specific antibodies in all species. Group mean total cholesterol (TC) concentration was reduced by up to 30%, and LDLc up to 50% in treated animals. Moreover, the PCSK9 vaccine-induced humoral immune response persisted for up to one year in mice, and reduced cholesterol levels significantly throughout the study. Finally, the vaccines were well tolerated in all species tested.

Conclusions: Peptide-based anti-PCSK9 vaccines induce the generation of antibodies that are persistent, high-affine, and functional for up to one year. They are powerful and safe tools for long-term LDLc management, and thus may represent a novel therapeutic approach for the prevention and/or treatment of LDL hypercholesterolemia-related cardiovascular diseases in humans.

Figure 1. Anti-PCSK9 vaccine efficacy in BALB/c mice.

(A) Antibody titers (ODmax/2) against huPCSK9 protein, for sera generated from different peptides generated upon 4 vaccinations in biweekly intervals. Values are means ±SEM (n = 8/group). Significance values relative to Peptide #1 (*p<0.05; **p<0.005, ***p<0.001) were obtained by one-way ANOVA test with subsequent Tukey’s multiple comparisons test. (B) Mean off-rate values ±SEM of antibodies generated upon three subsequent immunizations (marked by arrows) with Peptide #1 and original PCSK9 sequence in a biweekly interval: W0, W2 and W4. (C) PCSK9-LDLR binding affinities in the presence of anti-PCSK9 antibodies, generated from different peptides. Values are means ±SEM, shown as a percentage of that for the negative control. (D) LDLR levels in liver hepatocytes from mice inoculated with different peptides, shown as a percentage of the negative control. Pooled liver lysates within each group (n = 5/group) were analyzed; error bars represent mean values ±SEM of duplicate measurements. (E) Mean values of TC levels (mg/dL) for mice inoculated with different peptides. Values are means; error bars represent ±SEM. Samples for (A), (C), (D) and (E) taken at week 8. W: week.

Figure 2. Anti-PCSK9 vaccine efficacy in Wistar rats.

(A) Titers (ODmax/2) against huPCSK9 (n = 8/group) generated upon 4 vaccinations in biweekly intervals. (B) TC levels in treated animals (nmol/L), compared to control. Bars and error bars represent mean values ±SEM; significance compared to control values (****p<0.0001) was obtained by a 2-tailed Student’s t-test.

Figure 3. Vaccine-generated antibodies target plasma muPCSK9.

(A) muPCSK9 levels in plasma samples from pre-immunized (W0) and post-immunized mice (W8). Bars represent mean levels of detected muPCSK9 levels in plasma samples, error bars represent ±SEM, (n = 10 mice/group). Significance values were calculated by two-way ANOVA, followed by Bonferroni multiple comparisons test (n = 10 mice/group) (***p<0.001). (B) Direct detection of antibodies bound to plasma muPCSK9 in plasma samples from pre-immunized (W0) and post-immunized mice (W8). Increased OD₄₅₀ is indicative for anti-PCSK9-vaccine induced antibodies directly binding to muPCSK9. Bars represent mean values, error bars represent ±SEM, (n = 10 mice/group). Significance values were calculated by two-way ANOVA test with subsequent Bonferroni multiple comparisons test, (***p<0.001). W: week.

Figure 4. Long-term LDL cholesterol management upon anti-PCSK9 vaccination.

(A) Long-term evaluation over 50 weeks post prime immunization of titers against huPCSK9 (ODmax/2), generated upon 3 vaccinations in a biweekly interval (marked by arrows). Values are means ±SEM, n = 10 mice/group. (B) Measurement of TC values over the same time period as (A). Values are means; error bars represent ±SEMs. Significance values were calculated by comparison with the control group, by two-way ANOVA test, followed by Bonferroni multiple comparisons test, at the indicated time-points (n = 10 mice/group). Significance values are indicated as follows: *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001. (C) FPLC lipoprotein profile of pooled plasmas within the groups (n = 10), 10 weeks post prime immunization. (D) Plasma lipoprotein levels, for three classes (VLDL, LDL, HDL), 10 weeks after prime immunization, expressed as a percentage of that for the negative control. (E) FPLC cholesterol lipoprotein profiles (VLDL, LDL and HDL) of pooled plasmas within the groups (n = 10), 34 weeks post prime immunization. (F) Plasma lipoprotein levels, for three classes (VLDL, LDL, HDL), 34 weeks after prime immunization, expressed as a percentage of that for the negative control. W: week.

Figure 5. Anti-PCSK9 vaccine approach is boostable, and induced functional antibodies even after a single yearly re-boost.

(A) Off-rate values generated upon 3 vaccinations (marked with arrows) in a biweekly time period in BALB/c mice, followed by a single re-boost in week 54 (marked with an arrow), for samples (n = 10) taken at weeks 2, 4, 10, 30, 54 and 58. Values shown are means ±SEM. (B) R_max values (n = 10) for the vaccination timecourse shown in (A). Values shown are means ±SEM. Note the re-boost effect seen at week 58 on the amount of antibodies (R_max values) upon single re-vaccination in week 54. (C) Total cholesterol values for PCSK9-inoculated and control mice during a 58-week timecourse. Values are means ±SEM. Significance values were calculated by comparison with the control group, using a two-way ANOVA test with subsequent Bonferroni multiple comparisons test, at the indicated time-points (n = 10 mice/group). Significance values are indicated as follows: *p<0.05; **p<0.01; ***p<0.001. Note the slight loss of effect on TC levels over time (for example, by W42 and W50), and the induction of a significantly beneficial decrease following the re-boost immunization at W54. W: week.

Figure 6. Anti-PCSK9 vaccine significantly lowers the LDLc in Ldlr ^+/− male and female mice upon 4 immunizations.

(A) Mean titers (ODmax/2) ±SEM against huPCSK9 (n = 9). (B) TC levels (mg/dL), bars and error bars represent mean values (n = 9) ±SEM. (C) FPLC cholesterol lipoprotein profile (VLDL, LDL and HDL) of pooled plasmas from immunized mice (n = 9) and their related controls. (D) Mean values of the area under the curve (AUC) ±SEM evaluated by FPLC cholesterol lipoprotein profile analysis of the VLDL, LDL and HDL in immunized male mice and their related controls (n = 9). Significance values were calculated by comparison with control values for LDLc (****p<0.0001) using a 2-tailed Student’s t-test. (E) Mean titers (ODmax/2) ±SEM against huPCSK9 (n = 9). (F) TC levels (mg/dL), bars and error bars represent mean values (n = 9) ±SEM. (G) FPLC cholesterol lipoprotein profile (VLDL, LDL and HDL) of pooled plasmas (n = 9). (H) Mean AUC values ±SEM, evaluated by FPLC cholesterol lipoprotein profile analysis of VLDL, LDL and HDL cholesterol in immunized female mice and their related controls (n = 9). Significance values were calculated by comparison with negative control values, using a 2-tailed Student’s t-test. Significance values are indicated as follows: ****p<0.0001. All samples taken at week 8.

Figure 7. Anti-PCSK9 vaccination does not activate PCSK9-specific T cells.

ELISpot analysis for INF-γ-releasing splenocytes isolated from C57BL/6 (H-2D^b) mice. (A) naϊve mice (negative control). (B) mice immunized with KLH alone. (C) mice immunized with anti-PCSK9 vaccine. Triplicate measurements were performed from cell pools within the group. (D) Representative pictures from ELISpot assay are shown, including PHA stimulation as positive control. (E) ELISA analysis of anti-PCSK9 vaccine-induced antibodies against PCSK9 and IgE. The antibodies generated do not cross-react with IgE.